Dynamics of inertial spheroids in a decaying Taylor–Green vortex flow

Abstract

Inertial spheroids, prolates and oblates, are studied in a decaying Taylor–Green vortex (TGV) flow, wherein the flow gradually evolves from laminar anisotropic large-scale structures to turbulence-like isotropic Kolmogorov-type vortices. Along with particle clustering and its mechanisms, preferential rotation and alignment of the spheroids with the local fluid vorticity are examined. Particle inertia is classified by a nominal Stokes number [Formula: see text] which to first-order aims to eliminate the shape effect. The clustering varies with time and peaks when the physically relevant flow and particle time scales are of the same order. Low inertial ([Formula: see text]) spheroids are subjected to the centrifuging mechanism, thereby residing in stronger strain-rate regions, while high inertial ([Formula: see text]) spheroids lag the flow evolution and modestly sample strain-rate regions. Contrary to the expectations, however, spheroids reside in high strain-rate regions when the particle and flow time scales are comparable due to the dynamic interactions between the particles and the evolving flow scales. Moderately inertial ([Formula: see text]) prolates preferentially spin and oblates tumble throughout the qualitatively different stages of the TGV flow. These preferential modes of rotation correlate with parallel and perpendicular alignments of prolate and oblate spheroids, respectively, with the local fluid vorticity. However, for high inertial spheroids preferential rotation and alignment are decorrelated due to a memory effect, i.e., inertial particles require longer time to adjust to the local fluid flow. This memory effect is not only due to high particle inertia, as in statistically steady turbulence, but also caused by the continuously evolving TGV flow scales.