CFD modeling of single-phase flow in a packed bed with MRI validation

Computational fluid dynamics (CFD) simulations are becoming increasingly widespread with the advent of more powerful computers and more sophisticated software. The aim of these developments is to facilitate more accurate reactor design and optimization methods compared to traditional lumped‐paramete...

Full description

Saved in:
Bibliographic Details
Published in:AIChE journal 2012-12, Vol.58 (12), p.3904-3915
Main Authors: Robbins, David J., El-Bachir, M. Samir, Gladden, Lynn F., Cant, R. Stewart, von Harbou, Erik
Format: Article
Language:English
Subjects:
Applied sciences
Chemical engineering
Computational fluid dynamics
Exact sciences and technology
Fluid dynamics
Hydrodynamics of contact apparatus
magnetic resonance imaging
NMR
Nuclear magnetic resonance
Optimization techniques
preconditioning
reactor modeling
Reactors
Reynolds number
Simulation
trickle-bed reactor
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Computational fluid dynamics (CFD) simulations are becoming increasingly widespread with the advent of more powerful computers and more sophisticated software. The aim of these developments is to facilitate more accurate reactor design and optimization methods compared to traditional lumped‐parameter models. However, in order for CFD to be a trusted method, it must be validated using experimental data acquired at sufficiently high spatial resolution. This article validates an in‐house CFD code by comparison with flow‐field data obtained using magnetic resonance imaging (MRI) for a packed bed with a particle‐to‐column diameter ratio of 2. Flows characterized by inlet Reynolds numbers, based on particle diameter, of 27, 55, 111, and 216 are considered. The code used employs preconditioning to directly solve for pressure in low‐velocity flow regimes. Excellent agreement was found between the MRI and CFD data with relative error between the experimentally determined and numerically predicted flow‐fields being in the range of 3–9%. © 2012 American Institute of Chemical Engineers AIChE J, 2012
ISSN:0001-1541
1547-5905
DOI:10.1002/aic.13767