Session F05: Turbulence: Particle-laden Flows

Gravitatory settling of inertial particles in turbulent environments

Turbulent flows laden with particles are important for both environmental phenomena and industrial systems. This study focuses on dense sub-Kolmogorov inertial particles in a homogeneous and isotropic turbulent gas flow. Experimental measurements of particle vertical and horizontal velocities were taken via a Phase Doppler Interferometer. A wide range of Taylor Reynolds numbers (Re_λ ∈ [30-520]), Rouse numbers (Ro ∈ [0-5]) and volume fractions (φ_v ∈ [0.5×10^-5 - 2.0×10^-5]) were explored. Three types of grid turbulence were tested by means of a passive grid and an active grid used in two different modes (grid shafts moving randomly or open in a static configuration). This work aims at studying the role of the carrier phase Taylor Reynolds number, Rouse number and volume fraction on the settling velocity of inertial particles. We find, in agreement with previous work, that enhancement of the settling velocity occurs at low Rouse number, while hindering of the settling occurs at higher Rouse number for a decreasing turbulence energy levels (characterized by the velocity fluctuating intensity or the Taylor Reynolds number). The wide range of flow parameters explored allowed us to observe that enhancement decreases significantly with the Taylor Reynolds number.   

Particle extreme clustering and inward drifts measured at small separations in an isotropic turbulent flow

Recent experiments have shown explosive growth of the radial distribution function g(r) of particles in isotropic turbulence with r ^-6 scaling when r/η < 1 (r is particle-pair separation; η is Kolmogorov length scale), characteristic of hydrodynamic interaction. Using high-resolution particle tracking we measure g(r), inward radial relative velocities, and kinematic quantities for varying particle radius a and Stokes number at r down to near-contact (r/a = 2.07). A kinematic relation governing g(r) shows particles cluster when inward drift 〈w_r'(t)〉_r - ▽_rS₂ is < 0 (〈w_r'(t)〉_r is the mean particle-pair radial acceleration, -▽_rS₂ is the gradient of 2^nd-order velocity structure function). Extreme clustering regime statistics scale with r/a, indicating particle interactions. At the onset of extreme clustering,〈w_r'(t)〉_r turns strongly negative, indicative of inward-driving forces and increases magnitude as r decreases. After〈w_r'(t)〉_r magnitude peaks, -▽_rS₂ dominates. The experimental collision kernel estimate is O(10³)-O(10⁵) higher than simulations assuming non-interacting particles. Our study highlights the need to understand particle interactions responsible for extreme clustering.

  

Not Participating

Data from Direct Numerical Simulations of disperse bubbly flows in a vertical channel are used to study the effect of the bubbles on the carrier-phase turbulence. We developed a new method, based on an extension of the barycentric map approach, that allows us to quantify and visualize the anisotropy and componentiality of the flow at any scale. Using this we found that the bubbles significantly enhance anisotropy in the flow at all scales compared with the unladen case, and that for some bubble cases, very strong anisotropy persists down to the smallest scales of the flow. The strongest anisotropy observed was for the cases involving small bubbles. Concerning the energy transfer among the scales of the flow, our results indicate that for the bubble-laden cases, the energy transfer is from large to small scales, just as for the unladen case. However, there is evidence of an upscale transfer when considering the transfer of energy associated with particular components of the velocity field. Although the direction of the energy transfer is the same with and without the bubbles, the energy transfer is much stronger for the bubble-laden cases, suggesting that the bubbles play a strong role in enhancing the activity of the nonlinear term in the flow.   

Stokes point-particle dynamics and flow structure in stationary isotropic turbulence

Numerical simulation and (stochastic) closure modeling for the dynamics of inertial point particles in turbulence present many challenges beyond those encountered in Lagrangian fluid particle motion, even in regimes where the effects of finite particle size and two-way coupling may be considered small.  In this talk we examine a few fundamental aspects using direct numerical simulations of stationary isotropic turbulence at different Reynolds numbers. The Stokes number (defined as ratio of particle momentum response time to the Kolmogorov time scale) is varied from very small to very large. Datasets examined include the fluid velocity and velocity gradients (hence dissipation rate and enstrophy, which are highly intermittent) evaluated along the Stokes particle trajectories.  Numerical experiments involving filtering the turbulence at different scale sizes are also addressed.  The computations are performed using efficient parallel implementations whose costs are nearly independent of the particle count.   

Investigation of impinging sweeping jets through Particle Image Velocimetry

Sweeping jets are particular jets characterized by an oscillating motion. Such an oscillation is caused by the fluidic oscillator device, whose geometrical internal characteristics influence and govern this sweeping phenomenon. The most interesting characteristic of these devices is the absence of moving parts and piezo-electrical elements, making them a candidate for application in flow control and heat transfer fields. Many experimental studies and numerical simulations have focused on the application of sweeping jets for the control of air flow along aerodynamic surfaces while lower attention has been paid to their impinging flow field, fundamental to have a deep physical understanding of their heat transfer performances.

In order to characterize the mean flow field, the oscillating coherent flow field and the turbulent statistics of impinging sweeping jets phase-locked particle image velocimetry (PIV) measurements are carried out. Three fluidic oscillators devices have been studied. These devices are characterized by different mixing chamber lengths (2.5w, 3.5w and 4.5w). The issued sweeping jets have been investigated at a Reynolds number equal to 17,000 and for a nozzle-to-plate distance (H) ranging between 2 and 10 nozzle width (w).

The main results highlight that the fluidic oscillator mixing chamber length has a strong influence on the sweeping jet flow field in terms of time-averaged velocity components, coherent and uncoherent fluctuations distribution. In particular, this geometrical parameter also governs the presence or absence of the jet oscillating motion. In addition to that, the phase-locked measurements allow to reconstruct the oscillating motion of the jet flow field and the related turbulent fluctuations and to observe the presence of a vortex structure near the impingement plate moving coherently with the sweeping motion of the jet itself.   

Quantifying divergence and rotation of the inertial particle velocity in high Reynolds number turbulence using Voronoi and Delaunay tessellation

We propose finite-time measures to quantify the divergence and the curl of the velocity advecting point particle clouds in space and time. To this end we respectively determine the volume change rate and the rotation of cells using two subsequent time steps at the particle positions. We consider either Voronoi or Delaunay tessellation and assess the reliability of the two methods. We show a first order convergence in time and in space for divergence and curl for randomly distributed particles using Delaunay triangulation and a good agreement with the exact values. For the Voronoi tessellation we observe some off set in the case of randomly distributed particles due to geometrical effects. We apply these tools to three-dimensional direct numerical simulation data of particle-laden isotropic turbulence computed at high Reynolds number. We discuss and compare the results obtained with the two different techniques. For inertial particles the probability distribution functions (PDFs) of the divergence and of the curl deviate from that for fluid particles and we observe a similar Stokes number dependency for both tessellations. In the Delaunay case the extreme values of divergence and curl of the velocity are reduced and the corresponding PDFs are narrower. We also find different results in analyzing the mean divergence as a function of volume, which leads to different interpretations of the behavior of the particles as a function of scale.