Assessment of the point-wise approach for the Turbulent Settling of finite-size particles Journal: International Journal of Multiphase Flow
Francesco Battista, Sergio Chibbaro, Paolo Gualtieri · 2025
Read Assessment of the point-wise approach for the Turbulent Settling of finite-size particles Journal: International Journal of Multiphase Flow by Francesco Battista, Sergio Chibbaro, Paolo Gualtieri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-06481.txt.
Introduction
How fast do particles sink through turbulent water? Francesco Battista, Sergio Chibbaro, and Paolo Gualtieri test whether a cheap modeling shortcut gets the answer right. Their simulations treat each particle as a point, ignoring its size, for grains about ten times the smallest eddy scale and slightly denser than the fluid. They compare the results against fully resolved simulations that model every particle's surface.
The point model matched the resolved results well, including the reduced sinking speed seen for suspensions in water at a few percent concentration. Switching forces on and off in the equations showed what each contributes. When buoyancy dominates, turbulence barely matters and particles sink as if the water were still. As that dominance fades, turbulent motion takes hold and slows the fall. The authors argue that point models with careful force descriptions are effective tools for these suspensions.
Catalog record
| CREATOR | Francesco Battista, Sergio Chibbaro, Paolo Gualtieri |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-06481
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Assessment of the point-wise approach for the Turbulent Settling of finite-size particles Journal: International Journal of Multiphase Flow Francesco Battista, Sergio Chibbaro, Paolo Gualtieri Abstract We study the settling of suspensions of relatively large particles with a diameter of the order of ten Kolmogorov scales and density slightly larger than the carrier fluid in statistically steady homogeneous isotropic turbulence. The particle-to-fluid density ratio is varied to obtain a wide range of Galileo numbers, which are the ratios between buoyancy and viscous forces. We analyse the problem through high-resolved one-way coupled direct numerical simulations where the particles are modeled as material points. The physical parameters are chosen in the same range used in recent particle-resolved simulations (PRS, [ 1 , 2 ] ), against which we compare. The results of the point-wise simulations are in good agreement with the PRS ones, showing a reduced settling speed for the range of parameters under investigation, relevant to suspensions settling in aqueous media, at volume fractions up to a few percent for density ratios order of one. Results are obtained neglecting the inter-particles and particle-fluid interactions while purposely including/not including the different forces (e.g. Stokes drag, added mass, lift force) in the particles’ equations of motion to highlight their contributions respectively. At a high Galileo number, the mean settling velocity is only slightly affected by turbulent fluctuations, and it is the same obtained for the settling velocity of a single particle in a quiescent fluid. When the Galileo number is reduced, the settling velocity is progressively affected by turbulent fluctuations that cause a substantial decrease in the particle sedimentation speed. The transition occurs in a range of parameters where the settling velocity of the isolated particle in a quiescent fluid is the order of the root mean square value of the turbulent fluctuations. The present results are particularly relevant for applications. Point-wise models endowed with an accurate description of the hydrodynamic force are effective in capturing the particle settling speed and other higher-order statistics as demonstrated by direct comparison against particle-resolved simulations [ 1 , 2 ] . 1 Introduction Inertial particles are present in fluid flows in many engineering and environmental applications [ 3 , 4 , 5 , 6 , 7 , 8 ] . Among the most relevant, we can mention the transport of sediments, particle dispersion, granular flows, and volcanic eruptions. A key ingredient of all those processes is particle settling [ 9 , 10 , 11 ] , on which we focus in the present work. The dynamics depend on several physical quantities entering the experimental conditions, which may be represented by dimensionless parameters [ 12 ] . In particular, the particle-to-fluid density ratio \rho_{p}/\rho_{f} , the volume fraction of the suspension \phi_{V} , and the ratio between the buoyancy and the viscous forces given by the Galileo number Ga=\sqrt{(\rho_{p}/\rho_{f}-1)gd^{3}/\nu^{2}} , where g is the acceleration of gravity, d the diameter of the particle, and \nu the kinematic viscosity. Furthermore, settling may occur in a quiescent fluid or a turbulent flow. In turbulent flows, velocity fluctuations may significantly affect the settling velocity. In this case, it is also necessary to consider the Stokes number (the ratio between the particle and the flow time scales) and the Reynolds number. In isotropic conditions, it is practical to use the Taylor-scale Reynolds number Re_{\lambda}=u_{rms}\lambda/\nu , where u_{rms} is the root-mean-square of the velocity fluctuations and \lambda is the Taylor length-scale [ 13 ] . Despite many studies, the impact of turbulence on settling remains somewhat unclear. Given the several dimensionless parameters that control the phenomenon, intensive research is ongoing. Following an analysis of a simplified flow [ 14 , 15 ] , many Direct Numerical Simulations (DNS) and experiments provide evidence of an increase of the settling speed for small inertial heavy particles [ 16 , 17 , 18 , 19 ] , i.e. for particles with density ratios much larger than one and particle diameter smaller than the Kolmogorov scale. Enhancement of the settling velocity has also been found in detailed studies of laminar and turbulent 2D flows [ 20 , 21 ] . The dominant mechanism is a preferential sweeping leading to a bias in the trajectories. Other possible mechanisms for the settling enhancement may be the particle-particle hydrodynamic interactions [ 22 ] and two-way coupling effects [ 23 ] , both becoming important when increasing the volume fraction. Some of the same studies and other ones suggest that turbulence may inhibit settling when the control parameters are in specific ranges; for instance, when considering almost neutrally buoyant particles (particle-to-fluid density ratio order one) or relatively large particles (Stokes numbers order ten) [ 17 , 24 , 25 , 21 , 18 ] . Sometimes, controversial results have been reported. In particular, it has been claimed that the phenomenon of the reduction of the settling velocity may be observed in numerical simulations of heavy particles only when a nonlinear drag is used [ 18 ] , whereas a subsequent exhaustive study [ 19 ] reports always enhancement of the settling velocity both with linear and nonlinear drag. These studies underline that the problem of settling is rather complex and depends on many independent parameters. Moreover, results are sensitive to the large-scale flow structures, and particular attention has to be paid to the set-up and the subsequent numerical simulation [ 19 ] . The above studies focus on small heavy particles with an associate particle Reynolds number of order one. In this framework, only viscous drag and gravity play a role, and nonlinear corrections are negligible [ 19 ] . However, in relevant geophysical and hydraulic applications, water serves as the carrier fluid. Hence, for common materials, the particle-to-fluid density ratio is of order one, and the particle diameter is slightly larger than the Kolmogorov scale. In this range of parameters, all the different contributions to the hydrodynamic forces on the particles should be retrieved. However, the expression of the force is based on an asymptotic expansion [ 26 , 27 ] that may be a source of errors when particles have a diameter of the order of the Kolmogorov scale or larger, or the Reynolds number associated with particles is large. As seen above, the diameter of the particles is an important feature since it enters the volume fraction and the Galileo number, which controls the buoyancy effects. Moreover, for the sake of practicality, most of the studies on the subject have dealt with one-way coupling, neglecting the effects of the particles on the fluid. The one-way coupling point-wise approach is safe for particles smaller than the Kolmogorov scale and in dilute conditions [ 28 , 7 ] . Yet, when the number of particles is high and/or the diameter is large, the flow becomes semi-dilute or even moderately dense, with a dramatic impact on the flow [ 3 , 29 , 30 , 7 , 31 ] . On the other hand, even in dilute conditions, the interaction between particles larger than the Kolmogorov scale and the flow may lead to non-trivial phenomena, including long-range interactions, e.g. wakes. In these cases, both the point-wise and the one-way coupling approach may fail, and particle-resolved simulations (PRS) should be used [