Session E10: Particle-laden Flows: Particle-Turbulence Interaction I

Mean rise rate of oil droplets under breaking waves

Entrainment of oil slicks by breaking waves generates a spectrum of droplets. Knowledge of the droplet rise rate under wave motion and turbulence is essential for predicting the fate of oil spills. This experimental work investigates the trajectories of 20 µm to 1 mm droplets with a specific gravity of 0.86 under lab breaking waves. The background turbulence energy dissipation rate is ~1 m²s^-3 shortly after wave breaking and 10^-3 m²s^-3 60 s later. The droplet trajectories are recorded using cinematic digital inline holography. Data analysis follows the temporal evolution of the droplet rise rate and its impact on the size distribution of the remaining droplets as a function of the initial droplet statistics and evolution of the turbulence dissipation rate. Results show that the ratio of mean rise rate over the quiescent rise rate (V_s/V_q) increases with decreasing diameter consistent with Friedman and Katz (Phy. Fluids, 14, 2002). For droplets larger than 0.3 mm V_s/V_q <1, but for smaller droplets V_s/V_q >1. The strong correlation between V_s and the turbulent level suggests that preferential sweeping plays a role in enhancing V_s of small droplets. Incorporating these modifications in the rise rate due to the wave motion and turbulence is necessary to improve the oil spill models.   

The coupled effects between sediment and fluid in suspension layer during flow reversal under symmetric oscillatory sheet flow

Sediment transport under oscillatory sheet flow conditions provide an important example of strongly-coupled particle/turbulence interaction within an important practical flow. It has long been observed that concentration maxima occur at flow reversals, but they have not been well-captured in existing numerical simulations/models. A stereo PIV/PTV based multi-camera imaging method has been recently developed, enabling particle-resolved concurrent two-phase measurement in the middle of a typical sheet flow facility, with a thick mobile bed composed of well-sorted sediment particles (D₅₀=250μm, ρ_s=2.5g/cm³). Time-resolved highspeed recordings were conducted during flow reversals in a symmetric sinusoidal oscillatory sheet flow, with a period of 5s and excursion length of 0.80m. Within the suspended layer up to a volumetric fraction of close to 1%, time-series of 2D3C fluid velocities resolved down to sediment-scale were computed. In addition to each velocity field, instantaneous 3D sediment particle locations were reconstructed and tracked in consecutive frames to form trajectories. Derived statistics, such as sediment concentration, fluid and sediment kinematics and associated fluctuations, slip velocities etc. will be reported.   

Direct imaging of snow settling around turbulent eddies in the atmospheric surface layer

Turbulence-induced settling enhancement of inertial particles has been observed in the lab, simulations, and recent field measurements of snow particles (Nemes et al. JFM 2017; Li et al. JFM 2021). However, there is still no direct imaging of interaction between turbulent eddies and snow particles in the field. Here we present the first simultaneous field measurements of atmospheric flow using PIV (20 m wide x 40 m tall sample area) and snow particle trajectories using PTV (3 m x 5 m within the PIV domain). Such measurements show a preferential concentration of snow particles on the downward side of both retrograde and prograde vortices in the flow, with, in particular, the concentrated snow near the bottom of retrograde vortices. Moreover, snow particles are found to accelerate as they move toward the downward side of vortices and decelerate or even be lifted on the upward side. Additionally, we observe a stronger settling enhancement (i.e., a larger difference of settling velocity between downward and upward sides of vortices) by prograde vortices than that by retrograde vortices. Overall, our measurements provide direct field evidence and underlying physical processes of turbulence-induced snow settling enhancement in the atmospheric surface layer.   

Effects of large-scale flow structures on finite-size particle motion in turbulent boundary layers

While many studies focus on the interaction between particles with small inertia and near-wall structures in turbulent boundary layers, fewer have examined the effects of turbulence and the wall on larger particles. In the log region of a turbulent boundary layer, alternating slow- and fast-moving zones that dominate the flow field can alter particle dynamics. We perform 3D particle tracking and stereoscopic PIV measurements to study the translation and rotation of spheres with d⁺=56 and 116 (262<St⁺<1230) at Re_τ =670 and 1300 together with surrounding fluid velocities across streamwise-spanwise planes in the log layer. We compare spheres with specific densities of 1.006 and 1.152. Our results show that the particle velocity is strongly correlated to large-scale fluid motions. Wall friction impedes the denser sphere, causing it to lag the surrounding fluid, while the lighter sphere travels closer to the fluid velocity. In all cases, the spheres travel within both fast- and slow-moving zones, either consistently or alternately depending on their relative velocity. Vortex shedding is also prominent for the denser sphere, affecting the sphere velocity and altering the local turbulence.   

Modal analysis of particle laden turbulent flow in a square duct

Inertial particles in a turbulent flow have been shown to profoundly affect the morphology and intensity of coherent flow structures. Proper orthogonal decomposition (POD) is used in the present work to study the influence of particles on coherent flow structures by comparing unladen and laden flow results from two-way coupled DNS simulations with Lagrangian point-particle models. We exploit information on the contribution of most energetic structures to the overall turbulent kinetic energy (TKE) provided by POD to analyze turbulent modulation by particles. The study uses data from a fully developed turbulent flow in a square duct at a bulk Reynolds number of 20,000 with particle mass fraction ratios ranging from 0 to 1. The spatial modes identified by the POD show morphological differences with increasing particle loading. Selected terms from the turbulent kinetic energy budget and spectral analyses are used to help interpret the POD results. POD of the fluctuating particle concentration and velocity fields projected onto the velocity fluctuations modes allows analyzing correlations between the most energetic particle and flow structures.   

Mean Velocity Scaling in Dilute Particle-Laden Channel Flow at Moderate Mass Loading

In wall-bounded particle-laden turbulent flows with sufficiently high volume or mass loading, the mean velocity profile differs from a single-phase incompressible flow at the same Reynolds number. This alteration depends on the Stokes number and the mass loading of the particle phase, among other parameters. For a dilute mixture of heavy particles in a gas at moderate mass loading, we demonstrate how the mean fluid velocity and viscous stresses can be scaled using ideas from compressible flow transforms and rough-wall turbulent flows. This results in an improved collapse of mean velocity profiles, which we demonstrate with numerical simulations of particle-laden channel flow.

The suite of channel flow simulations were conducted at Re_τ≈235, using Lagrangian point particles, with a range of Stokes numbers (St⁺≈1-60) and particle mass loadings (Φ =10%-300%).   

Particle capture by big drops in turbulent channel flow

We examine the process of particle capture by large deformable drops in turbulent channel flow. We simulate the solid-liquid-liquid three-phase flow with an EulerianLagrangian method based on direct numerical simulation of turbulence coupled with a phase-field model, to capture the interface dynamics, and Lagrangian tracking of small (sub-Kolmogorov) particles. Drops have the same density and viscosity of the carrier liquid, and neutrally buoyant, quasi-inertialess, solid particles are one-way coupled with the other phases. Our results show that particles are transported towards the interface by jetlike turbulent motions and, once close enough, are captured by interfacial forces in regions of positive surface velocity divergence. These regions appear to be well correlated with high-enstrophy flow topologies that contribute to enstrophy production via vortex compression or stretching. Examining the turbulent mechanisms that bring particles to the interface, we have been able to derive a simple transport model for particle capture. The model is based on a single turbulent transport equation in which the only parameter scales with the turbulent kinetic energy of the fluid measured in the vicinity of the drop interface, and its predictions of the overall capture efficiency agree remarkably well with numerical results.   

Interface topology and evolution of particle patterns on large deformable drops in turbulence

The capture of neutrally-buoyant, sub-Kolmogorov particles at the interface of large deformable drops in turbulent flow and the subsequent evolution of particle surface distribution are investigated. Direct numerical simulation of turbulence, phase field modeling of the drop interface dynamics and Lagrangian particle tracking are used. Particle distribution is obtained considering excluded-volume interactions, i.e. by enforcing particle collisions. Particles are initially dispersed in the carrier flow and are driven in time towards the surface of the drops by jet-like turbulent fluid motions. Once captured by the interfacial forces, particles disperse on the surface. Excluded-volume interactions bring particles into long-term trapping regions where the average surface velocity divergence sampled by the particles is zero. These regions correlate well with portions of the interface characterized by higher-than-mean curvature, indicating that modifications of the surface tension induced by the presence of tiny particles will be stronger in the highly-convex regions of the interface.   

Not Participating

Bubble--particle collisions in turbulence are key to engineering processes. However, the behaviour of suspended species in turbulent flow fields are intricate and sometimes counter-intuitive, making it difficult to predict the collision rate. For example, bubbles and particles segregate as they preferentially concentrate in different flow regions. While this effect is expected to reduce the effective inter-species collision rate, enhanced relative velocities due to opposite responses to fluid accelerations may counteract this phenomenon. This study aims to unravel how these potentially competing effects influence the collision rate of bubbles and particles in turbulence. We perform simulations using the point-particle approach to cover a range of the relevant parameters, such as the Stokes number and the Taylor Reynolds number. Our results suggest existing models in the literature do not capture the collision rate in the tested parameter range accurately. This highlights the need for models that more truthfully capture the underlying physics.