Three‐Dimensional Computational Fluid Dynamics Modeling of a Prismatic Spouted Bed

Abstract

Modern simulation tools like discrete particle models or continuum models have been established as precious methods to gain insight into the complex processes in fluidized and spouted beds. Valuable information on parameters influencing the hydrodynamics can be obtained directly from computer simulations. Numerous researchers applied two-dimensional discrete and continuum models to simulate fluidized and spouted beds. However, only few publications deal with three-dimensional continuum simulations, and very often results of three-dimensional experiments are compared with two-dimensional results. In this work, a three-dimensional simulation by application of a continuum model is presented in a prismatic spouted bed with slit-shaped gas inlets. Comparisons with two-dimensional simulations in the same geometry (Gryczka et al. 2009, Chem. Eng. Sci., in press) are also part of this publication. The influence of the third dimension on the hydrodynamic behavior (bubble formation, bubble size, etc.) will be discussed.

Characterization and CFD-DEM modelling of a prismatic spouted bed

“…Therefore, the process parameters obtained with a lab scale prismatic apparatus, cannot be applied (at least one-to-one) for large plants, if the apparatus is designed too small. However, according to the results in [22], this influence and threedimensional behaviour decrease, if the distance between vertical walls is high.…”

“…Regarding all three criteria in combination, a reasonable resemblance was achieved by the oldest of the used drag correlations; however the bed expansion and velocities of particles in comparison with the PIV results were underpredicted. A 3D-CFD investigation (Euler-Euler-approach) was done by Gryczka et al [22] to study the influence of the apparatus depth on the prevailing gas-solid hydrodynamics. An apparatus with the same plane geometry as investigated experimentally [1] and modelled in 2D [20,21], and a high extension in the depth direction was simulated.…”

“…Thus, the discrepancy between 2D simulations and PIV measurements reported in [20] (under-predicted bed expansion and particle velocities) can be probably explained with the help of a 3D simulation, since the bed expansion and the particle velocities are higher in the vicinity of the vertical walls than in the centre of the apparatus, which does not appear in a 2D model. However, another 3D CFD study [22] carried out on this apparatus to investigate the influence of the third dimension, by means of a model with a significantly higher depth of 400 mm (α/β = 0.625), showed only minor variations in the bed height, particle velocities and pressure behaviour along the depth direction. Therefore it was assumed, that the influence of the depth dimension on the flow structure is quite low and that the particle dynamics in this apparatus can be assumed to be independent on the depth [22,40].…”

“…However, another 3D CFD study [22] carried out on this apparatus to investigate the influence of the third dimension, by means of a model with a significantly higher depth of 400 mm (α/β = 0.625), showed only minor variations in the bed height, particle velocities and pressure behaviour along the depth direction. Therefore it was assumed, that the influence of the depth dimension on the flow structure is quite low and that the particle dynamics in this apparatus can be assumed to be independent on the depth [22,40]. However, this DPM study shows that this is not always the case and a low distance between vertical walls can cause three-dimensional behaviour (at least in dense regimes).…”

Estudio de transporte de momento y masa en un biorreactor de partición de tres fases (agua-aire-líquido iónico), utilizando dinámica de fluidos computacionales (CFD), para escalamiento descendente

“…El modelado a nivel local de los diferentes fenómenos de transporte se basa en las aportaciones numéricas de John Von Neumann el cual estudió la estabilidad numérica en sistemas dinámicos (Krasnopolskaya et al, 1999). Por otro lado, Clough (1960) propuso el uso de un método numérico llamado elementos finitos, el cual es utilizado actualmente en la mayoría de los simuladores, con aplicación en el análisis estructural (Donea, 1992) realizan simulaciones tridimensionales en sistemas con geometría sencillas que no presentan problemas de convergencia (Gryczka et al, 2009).…”