The modulation of decaying isotropic turbulence by 45 000 spherical particles of Kolmogorov-length-scale size is studied using direct particle–fluid simulations, i.e. the flow field over each particle is fully resolved by direct numerical simulations of the conservation equations. A Cartesian cut-cell method is used by which the exchange of momentum and energy at the fluid–particle interfaces is strictly conserved. It is shown that the particles absorb energy from the large scales of the carrier flow while the small-scale turbulent motion is determined by the inertial particle dynamics. Whereas the viscous dissipation rate of the bulk flow is attenuated, the particles locally increase the level of dissipation due to the intense strain rate generated near the particle surfaces due to the crossing-trajectory effect. Analogously, the rotational motion of the particles decouples from the local fluid vorticity and strain-rate field at increasing particle inertia. The high level of dissipation is partially compensated by the transfer of momentum to the fluid via forces acting at the particle surfaces. The spectral analysis of the kinetic energy budget is supported by the average flow pattern about the particles showing a nearly universal strain-rate distribution. An analytical expression for the instantaneous rate of viscous dissipation induced by each particle is derived and subsequently verified numerically. Using this equation, the local balance of fluid kinetic energy around a particle of arbitrary shape can be precisely determined. It follows that two-way coupled point-particle models implicitly account for the particle-induced dissipation rate via the momentum-coupling terms; however, they disregard the actual length scales of the interaction. Finally, an analysis of the small-scale flow topology shows that the strength of vortex stretching in the bulk flow is mitigated due to the presence of the particles. This effect is associated with the energy conversion at small wavenumbers and the reduced level of dissipation at intermediate wavenumbers. Consequently, it damps the spectral flux of energy to the small scales.
Direct particle–fluid simulations of heavy spheres and ellipsoids interacting with decaying isotropic turbulence are conducted. This is the rigorous extension of the spherical particle analysis in Schneiders et al. (J. Fluid Mech., vol. 819, 2017, pp. 188–227) to $O(10^{4})$ non-spherical particles. To the best of the authors’ knowledge, this represents the first particle-resolved study on turbulence modulation by non-spherical particles of near-Kolmogorov-scale size. The modulation of the turbulent flow is precisely captured by explicitly resolving the stresses acting on the fluid–particle interfaces. The decay rates of the fluid and particle kinetic energy are found to increase with the particle aspect ratio. This is due to the particle-induced dissipation rate and the direct transfer of kinetic energy, both of which can be substantially larger than for spherical particles depending on the particle orientation. The extra dissipation rate resulting from the translational and rotational particle motion is quantified to detail the impact of the particles on the fluid kinetic energy budget and the influence of the particle shape. It is demonstrated that the previously derived analytical model for the particle-induced dissipation rate of smaller particles is valid for the present cases albeit these involve significant finite-size effects. This generic expression allows us to assess the impact of individual inertial particles on the local energy balance independent of the particle shape and to quantify the share of the rotational particle motion in the kinetic energy budget. To enable the examination of this mechanistic model in particle-resolved simulations, a method is proposed to reconstruct the so-called undisturbed fluid velocity and fluid rotation rate close to a particle. The accuracy and robustness of the scheme are corroborated via a parameter study. The subsequent discussion emphasizes the necessity to account for the orientation-dependent drag and torque in Lagrangian point-particle models, including corrections for finite particle Reynolds numbers, to reproduce the local and global energy balance of the multiphase system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.