Numerical prediction of Péclet number in small-sized fixed bed reactors of spheres. (31st August 2021)

• Record Type: Journal Article
• Authors: Petrazzuoli, Vittorio; Rolland, Matthieu; Sassanis, Vasileios; Ngu, Vincent; Schuurman, Yves; Gamet, Lionel
• Is Part Of: Chemical engineering science. Volume 240(2021)
• Publication Date: 2021-08-31
• Subjects: Computational Fluid Dynamics -- Small fixed beds of spheres -- Axial Péclet number -- Direct Numerical Simulations -- Laminar flow -- Moments of age; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2021.116667
• Languages: English
• ISSNs: 0009-2509

Abstract

Graphical abstract: Highlights: DEM-OpenFOAM® workflow to compute Pe number in packed beds of spheres in laminar flow. Workflow faster and more informative than experiments. Complex influence of the reactor-to-sphere diameter ratio on Pe number. Pe number uncertainty decreases as a function of the reactor length to the power of −0.5. Abstract: Fluid dispersion in small fixed bed reactors under single phase laminar flow conditions is evaluated through accurate predictions of axial Péclet numbers Peax using Direct Numerical Simulations (DNS). The fixed bed packing is obtained through the Discrete Element Method (DEM) code Grains3D. Peax numbers are computed from steady state simulations of the flow. An additional set of equations solving the spatial distribution of the moments of the age distribution is used. This gives access to local information on dead zones or bypasses. Computations are realized with the OpenFOAM® open-source library. Results are close to the experimental data. The DEM-OpenFOAM® workflow benefits from an accurate control of the fluid physical properties and packing geometry, thus leading to a faster production of data than an experimental approach. We first explore the influence of the reactor-to-sphere diameter ratio δ ( δ < 4), fluid velocity and molecular diffusivity on the Peax number. The behavior proves much more complex than any previous report, and it can be qualitatively explained by looking at the packing structure. Then, we use the DEM-OpenFOAM® workflow to quantify the uncertainty on the Peax number when repeating the packing of a δ = 4 fixed bed reactor, and we propose that the uncertainty on the Peax number decreases as a function of the reactor length to the power of −0.5.