(626a) Turbulence and Reaction Modeling in Particle-Resolved Packed-Bed Reactors: What Level of Detail Is Needed?

Packed-bed reactors are widely used amongst others for highly endothermic or exothermic catalytic reactions. These reactor types are characterized by the complex interplay between flow, species and temperature distribution which is affected by the catalyst loading as well as the shape of the catalytic particles. All of them are affecting the local reaction rates, and, therefore, the yield as well as the selectivity. Since the experimental accessibility is limited and not trivial, numerical simulations open the door to a detailed understanding and optimization of such reactors. This was achieved due to substantial advances in the last two decades in modeling particle-resolved packed-beds [1,2,3], allowing a detailed insight into the flow, species, and temperature distribution within the beds, and, therefore, also into the local conversion of the surface reactions. Based on these simulations, model parameters for simplified 1D-models can be estimated[4] and these simplified models can then be used to calculate a full multi-tubular reactor by coupling CFD simulations with a flow-sheet or process simulation tool[5].

In this contribution two unanswered questions are addressed. Although there are various turbulence models available, most of them are derived for external flows and their applicability to porous structures is not known. To address this, DNS simulations are conducted, using a unit-cell approach, and the results are compared with simulations that use different RANS turbulence models, like k-epsilon and k-omega in terms of runtime and accuracy. The second question is related to the spatially resolved simulation of diffusion and reaction in porous particles itself. For this we show a new approach in Simcenter STAR-CCM+ which allows an accurate species and temperature transfer between the fluid and the particles, diffusion, and reaction inside the particle, as well as a full suppression of any externally initiated convective flow.

1. https://doi.org/10.1016/j.cej.2010.10.053
2. https://doi.org/10.1016/j.ces.2014.09.007
3. https://doi.org/10.1515/revce-2017-0059
4. https://doi.org/10.3390/en14102913
5. https://doi.org/10.3303/CET2186141