Authors: Casey Q. LaMarche a, Ben Freireich b, Ray A. Cocco c, Jia Wei Chew d
a Particulate Solid Research, Inc., 4201 W. 36th Street, Chicago, IL 60632, USA
b Origin Materials, 930 Riverside Pkwy, Suite 10, West Sacramento CA 95605, USA
c Particles in Motion, LLC, Elmhurst, IL 61026 USA
d School of Chemical and Biomedical Engineering, Nanyang Technological University, 62 Nanyang Dr, Singapore 637459, Singapore
Source: This paper was published in Chemical Engineering Journal.
Abstract: Numerical methods like computational fluid dynamics (CFD) for predicting multiphase flow involving solid particles offer promising benefits to a wide range of applications. Validated predictions of fluidization behavior would benefit such industrially relevant applications. Accurate representation of the fluid-particle momentum exchange, i.e., the drag and buoyancy forces, is one key to developing reliable mathematical tools for studying fluid-particle hydrodynamics. A significant number of drag correlations exist, yet little (widely accepted) guidance is available on which formulation is the most appropriate. The contributions focusing on predicting the drag from flows through random assemblies of monodisperse spheres alone comprises a substantial subset of the available literature. An assumption of homogeneous distribution of solids is at the core of an overwhelming majority of these drag correlations, i.e., homogeneous drag correlations. Hence, we review correlations for predicting the drag force for flows past homogeneous suspensions of spherical particles with narrow size distributions, which are the foundation for (i) predicating the drag force of flow past non-ideal particle assemblies (non-spherical, polydisperse, etc.), and (ii) developing sub-grid drag closures. Here, many of the commonly used homogeneous drag models are overviewed in detail, including the datasets and methods for development and validation. Potential sources of error between model prediction are identified and the accumulation of uncertainty in model development is presented. We demonstrate a lack of drag model development and/or validation over the complete parameter space (solids volume fractions, Reynolds numbers, density ratios, etc.) relevant to industrial fluid-particle flows, and discuss the underpinning physical and computational limitations driving this gap.