Dissecting inertial clustering and sling dynamics in high-Reynolds number particle-laden turbulence

Abstract: In this work, we aim to deepen the understanding of inertial clustering and the role of sling events in high-Reynolds number ($Re$) particle-laden turbulence. To this end, we perform one-way coupled particle tracking in flow fields obtained from direct numerical simulations (DNS) of forced homogeneous isotropic turbulence. Additionally, we examine the impact of filtering utilized in large eddy simulations (LES) by applying a sharp spectral filter to the DNS fields. Our analysis reveals that while instantaneous clustering through the centrifuge mechanism explains clustering at early times, the path history effect--the sampling of fluid flow along particle trajectories--becomes important later on. The filtered fields expose small-scale fractal clustering that cannot be predicted by the instantaneous flow field. We show that there exists a filter-effective Stokes number that governs the degree of fractal clustering and preferential sampling, revealing scale-similarity in the spatial distributions and fractal dimensions. Sling events are prevalent throughout our simulations and impose prominent patterns on the particle fields. In pursuit of investigating the sling dynamics, we compute the relative velocity, ensemble-averaged over proximal neighboring particles, to identify particles undergoing caustics. As postulated in recent theories, we find that in fully resolved, high-$Re$ turbulence, sling events occur in thin sheets of high strain, situated between turbulent vortices. This behavior is driven by rare, extreme events of compressive straining, manifested by fluctuations of the flow velocity gradients that propagate back and forth the positive branch of the Vieillefosse line.