Effect of particle shape on fluid statistics and particle dynamics in turbulent pipe flow

A Gupta,F Toschi,H J H Clercx

doi:10.1140/epje/i2018-11724-6

Abstract

.Anisotropic particles are present in many natural and industrial flows. Here we perform direct numerical simulation (DNS) of turbulent pipe flows with dispersed finite-size prolate spheroids simulated by means of the lattice Boltzmann method (LBM). We consider three different particle shapes: spheroidal (aspect ratio 2 and 3) and spherical. These three simulations are complemented with a reference simulation of a single-phase flow. For the sake of comparison, all simulations, laden or unladen have the same energy input. The flow geometry used is a straight pipe with length eight times its radius where the fluid is randomly seeded with 256 finite-size particles. The volume fraction of particles in the flow has been kept fixed at 0.48% by varying the major and minor axis of each particle such that their volume remains the same. We studied the effect of different particle shapes on particle dynamics and orientation, as well as on the flow modulation. We show that the local accumulation of spheres close to the wall decreases for spheroids with increasing aspect ratio. These spheroidal particles rotate slower than spheres near to the wall and tend to stay with their major axes aligned to the flow streamwise direction. Despite the lower rotation rates, a higher intermittency in the rotational rates was observed for spheroids and this increase at increasing the aspect ratio. The drag reduction observed for particles with higher aspect ratio have also been investigated using the one-dimensional energy and dissipation spectra. These results point to the relevance of particle shapes on their dynamics and their influence on the turbulent flow.Graphical abstract

Highlights

The study of solid particle suspensions in turbulence is relevant for the understanding and for the optimization of many biological and engineering flows [1]
Non-spherical particles, with varying shapes and sizes are common: plankton species and pollen grains occur in an astonishing variety of shapes, cellulose fibers or textile vary in their rigidity and shape, microalgae in photo-bioreactors are usually anisotropic in shape [2]
The simulation algorithm consists of three major components: the fluid solver, the model to simulate the dynamics of the particles and the particle-particle/particle-wall interaction

Summary

Introduction

The study of solid particle suspensions in turbulence is relevant for the understanding and for the optimization of many biological and engineering flows [1]. A comprehensive review of the models used to describe non-spherical particle motion, along with numerical and experimental methods for measuring particle dynamics, has been provided in [14] The majority of these previous studies, mostly focusing on turbulence, limited to the case of point-like spheroidal particles, assumed dilute conditions, and neglected the feedback on the flow. Researchers investigated, by means of DNS, the transport and the deposition of ellipsoidal particles in a turbulent channel flow [37] They provided velocity and orientational particle statistics in the viscous sublayer and in the buffer layer. In [42] researchers studied the effects of particle inertia, particle shape, and fluid shear on particle rotation using the direct numerical simulation of the turbulent channel flow.

Methodology

Flow geometry

Fluid phase analysis

Particle dynamics

Findings

Conclusions