401012 Is CFD Ready for Being a Design Tool for Fixed-Bed Reactors?

Tuesday, April 28, 2015: 1:30 PM
416AB (Hilton Austin)
Gregor D, Wehinger, Thomas Eppinger and Matthias Kraume, Chair of Chemical and Process Engineering, Technische Universität Berlin, D-10587 Berlin, Germany

Is CFD ready for being a design tool for fixed-bed reactors?

Gregor D. Wehinger, Thomas Eppinger, Matthias Kraume

Chemical and Process Engineering, Technische Universität Berlin, Fraunhoferstr. 33-36, 10587 Berlin, Germany

            Almost 15 years ago Dixon and Nijesmeisland simulated for the first time a fixed bed with a small tube-to-particle-diameter ratio (N) that took the actual particle shape into account. Although the bed consisted of only some uniformly packed spheres, it was now possible to investigate the interstitial flow field without a priori assumptions for momentum transport. Consequently, the authors claimed “CFD as a design tool for fixed-bed reactors” [1]. After that groundbreaking paper, several scientists have extended these detailed CFD simulations toward non-spherical shaped particles, particles including thermal conductivity, as well as catalytic active particles. However, several aspects of an adequate modeling are still under investigation, i.e., among others turbulent flow, meshing criteria, radiation, and the coupling of heterogeneous catalysis on the particle surface with the surrounding flow field.

            In this work we are looking critically at recent developments in the field of spatially resolved simulations of catalytic fixed-bed reactors. In addition, examples are given, how such simulations can be validated with appropriate detailed experimental techniques. Therefore, a capillary sampling technique is applied in a tubular fixed-bed reactor [2]. This method provides profiles of gas species concentrations, gas temperature and surface temperature. These experimental results are then compared with 3D numerical simulations taking explicitly into account the shape of the spherical particles. Inside the particles heat conduction is assumed. On the particle surface a detailed reaction mechanism is applied distinguishing between adsorption, surface reaction and desorption of chemical species. The dry reforming of methane (DRM) is taken as an example of heterogeneous catalytic processes. The procedure of the fixed-bed model follows that of Wehinger et al. [3]. This modeling approach allows the investigation of the interactions between chemical kinetics and transport of momentum, heat and mass.

            Finally the applicability of CFD as a design tool is investigated. Therefore, three different kinds of particles of typical industrial dimensions are evaluated under common reaction conditions toward conversions and yields of the DRM. The particle shapes involve spheres, cylinders and 1-hole cylinders, see Fig. 1. With such detailed modeling transport phenomena inside fixed-beds can be quantified and finally, a substantial comparison between different packings is possible.

Fig. 1. Surface site fraction of adsorbed carbon C(S) on cylindrical particles (left) and 1-hole cylindrical partciles (right) for DRM.

            However, the time required for a converged simulation can account for several days. Accordingly, such simulations are too time-consuming that they can replace lumped models, which deliver results in seconds, in the design stage. Nevertheless, detailed simulations of small N fixed-bed reactors offer unique insights into the complex processes of catalytic reacting flow. Moreover, dependencies on empirical equations can be minimized.

References

[1]       Dixon, A. G. & Nijemeisland, M. Ind. Eng. Chem. Res., 2001, 40, 5246-5254

[2]       Horn, R., Degenstein, N., Williams, K. & Schmidt, L. Catal. Lett., 2006, 110, 169-178

[3]       Wehinger, G.D., Eppinger, T. & Kraume, M. Chem. Eng. Sci., 2015, 122, 197-209

 


Extended Abstract: File Not Uploaded