Investigation of particle clustering in unsteady wake flows using Direct Numerical Simulations (DNS)

The flow past a bluff body, such as a circular cylinder, exhibits flow separation and the formation of a periodic von Kármán vortex street in the wake, characterized by interactions between the boundary layer, separating fluid shear layers, and unsteady wake dynamics. These flow dynamics are highly sensitive to the Reynolds number (Re) and are further influenced by the presence of suspended particles. In the present study, numerical simulations are conducted using DNS to investigate the interaction between inertial particles and vortical structures generated downstream of a cylinder in an unbounded domain. Voronoi tessellation methodology is employed to analyze non-uniform spatial clustering of particles by identifying clusters formed under the influence of vortical structures. The present analysis is carried out to study the effects of Reynolds number, Stokes number (St), and particle seeding on the vortex shedding. It was observed that increasing both St and particle seeding in the fluid leads to a suppression of vortex shedding and a reduction in drag experienced by the cylinder. Furthermore, particles at higher Stokes numbers tend to accumulate and cluster upstream of the cylinder due to particle-cylinder collisions, forming a bow shock-like structure. The evolution of Karmen vortices were tracked downstream to the cylinder, revealing that vortex dissipation is strongly influenced by the particle Stokes number and the inlet particle seeding rate. These findings highlight the need for the study of complex two-way coupling between fluid and particle phase dynamics and their influence on unsteady wake flows in particle-laden multiphase flow.