Clustering in laboratory and numerical turbulent swirling flows

TL;DR

This study investigates how various flow geometries influence the clustering behavior of flow features and inertial particles in turbulent flows, combining simulations and experiments to reveal consistent particle clustering patterns across different flow topologies.

Contribution

It demonstrates that Taylor-scale neutrally buoyant particles cluster similarly to inertial particles across different turbulent flow geometries, supported by combined numerical and experimental analysis.

Findings

Flow topology affects clustering of flow nulls but not inertial particles.

Neutrally buoyant particles exhibit clustering behavior similar to inertial particles.

Clustering patterns are consistent across different flow geometries.

Abstract

We study the three-dimensional clustering of velocity stagnation points, of nulls of the vorticity and of the Lagrangian acceleration, and of inertial particles in turbulent flows at fixed Reynolds numbers, but under different large-scale flow geometries. To this end, we combine direct numerical simulations of homogeneous and isotropic turbulence and of the Taylor-Green flow, with particle tracking velocimetry in a von K\'arm\'an experiment. While flows have different topologies (as nulls cluster differently), particles behave similarly in all cases, indicating that Taylor-scale neutrally buoyant particles cluster as inertial particles.