Session Y05: Particle-Laden Flows: Non-Spherical Particles (11:30am - 12:15pm CST)

Particle-fluid-wall interaction of anisotropic inertial particles in a turbulent boundary layer

The dynamics of dilute, inertial rod- and disk-shaped particles in a saltation-suspension regime are studied in a turbulent boundary layer. Simultaneous, time-resolved flow fields, particle trajectories, and particle orientations are obtained using particle image velocity and particle tracking velocimetry in a paddlewheel-driven water channel. Statistics of particle velocity, particle acceleration, and fluid velocity interpolated at particle locations are computed to investigate particle interaction with the fluid turbulence and with the wall. Both types of particle interact with the wall by touching down, sliding, rolling or tumbling, then lifting off. Particle lift-off is strongly associated with fluid ejection events occurring at the particle location, while touchdown is only weakly associated with sweep events. Particle orientation is compared between particle types and is found to differ significantly. Rod particles preferentially orient their axis of rotational symmetry in the streamwise direction and not in the vertical, whereas disks preferentially orient their symmetry axis in the vertical direction and not in the streamwise. In both cases the symmetry axis is rarely oriented spanwise, i.e., rarely aligned with the mean vorticity. These preferences are strongest in the near-wall region. Tumbling behavior is observed for both particle types, especially near the wall, and strongly affects the particle-fluid-wall interplay compared to spherical particles of similar inertia.   

A physics-Based Statistical Model for Nanoparticle Deposition

Nano particle deposition is widely observed in biological systems, medical procedures, and manufacturing processes, etc. However, while instructive, the theoretical studies found in the literature only describe the phenomena on a case-by-case basis without a unifying framework. Physically, whether a nanoparticle will deposit on a substrate solely depends on the substrate-particle-solution interfacial energies and the particle's incident velocity vector upon impact. Since the number of particles is usually too big to consider every single one of them, much of this phenomenon is stochastic in nature. In this paper, a physics-based mathematical model is proposed to analyze nanoparticle deposition on a substrate. Based on statistical and physical considerations, the model quantitatively predicts the substrate's coverage evolution and growth rate. Its validity was verified by a dip coating experiment where a Polydimethylsiloxane (PDMS) substrate was periodically immersed in a sonicated graphene nanosheets solution. This study is expected to spur future endeavors in systematically characterizing film coating, drug delivery and other processes involving particle depositions.   

Effects of rarefaction and shape on the drag of spheroidal particles using the DSMC method

The capability to simulate a two-way coupled interaction between a rarefied gas and an arbitrary-shape colloidal particle is important for many practical applications. By means of numerical simulations based on the Direct Simulation Monte Carlo (DSMC) method we investigate the influence on the drag coefficient in case of ellipsoidal particles of several parameters, including rarefaction conditions, the particle orientation, the particle aspect-ratio, and the gas-solid boundary conditions. The motivation is to model the transport of such particles in high tech mechanical systems. We evaluate the drag force exerted by the gas on the ellipsoidal particle using a momentum exchange approach for an oblate and prolate ellipsoid, at different angles of attack and rarefaction levels. The drag force is then compared with the one obtained for a spherical particle at equivalent rarefaction conditions.   

Behavior of settling ellipsoidal microparticles in homogeneous isotropic turbulence

Direct numerical simulations were conducted for the motion of heavy ellipsoidal particles in isotropic turbulence for an investigation of the effect of gravity on the behavior of ellipsoidal microparticles. For various Stokes numbers and gravity factors, we explore how particle’s rotation depends on fluid structures and gravitational force. The particle rotation comprises tumbling (rotating about the axis of rotation of a particle) and spinning (rotating perpendicular to tumbling). When the aspect ratio of the particle is low, the ratio of tumbling is large, and as the aspect ratio increases, the ratio of spinning increases. We found that gravity strongly affects the trajectory and the behavior of rotation of ellipsoidal particles. The preferential clustering pattern of the heavy ellipsoidal particles was different from that of spherical particles depending on the Stokes number and the aspect ratio of the particles.   

Fall characteristics of snowflakes in clouds: settling of complex non-spherical particles in quiescent fluid

The fall characteristics of ice particles in clouds is of fundamental importance to simulating their microphysical evolution in numerical models, and to interpretation of in-situ and remote-sensing observational data. Yet remarkably little is understood about this problem. While much is known about how spheroids, discs etc sediment through fluids there is a critical lack of data about the types of particle which prevail in the atmosphere where irregular polycrystals and aggregates are the norm. To solve this problem we present results from a comprehensive series of experiments where analogues of natural snowflakes have been created using a 3D printer, and dropped in glycerine solutions. The falling analogues were tracked in 3D space over time using multiple cameras, and their orientation and trajectory reconstructed digitally. We find that many irregular particles are (despite their high complexity) extremely stable, adopting a preferred orientation (which, unexpectedly, is often Reynolds-number dependent), but rotate steadily around a vertical axis as they fall, tracking out a spiral trajectory and how the drag is sensitive to porosity. We also present data from new experiments using volumetric PIV to retrieve 3D velocity fields, providing further insight into the problem.   

Alignment of infinitesimal material lines and surfaces in wall turbulence

An infinitesimal material line or surface are both referred to as an infinitesimal material element, which is defined as an object always consisting of the same group of fluid particles and passively following the fluid motion. We numerically investigate the Lagrangian evolution of infinitesimal material elements in a turbulent channel flow with a focus on the interaction between turbulent coherent structures and the material elements. Based on the ensemble-averaged coherent vortices and the alignment pattern of material elements, we classify three regions from the wall to the center of channel, i.e. shear-dominance region, structure-dominance region and isotropic region. We find that each region is characterized by its distinct alignment pattern and alignment mechanism. Our work highlights the importance of coherent structures on the alignment behavior of material elements in the structure-dominance region and proposes an alignment mechanism, which has the implications in the relevant studies including polymers or fiber induced drag reduction of wall-turbulence and orientation distribution of wood fibers in paper making process.   

Experimental investigation of disks settling in homogeneous air turbulence

The dynamics of non-spherical particles falling through turbulent air is directly relevant to atmospheric precipitation. The detailed interaction with the surrounding fluid has largely been studied in water due to reduced requirements in spatio-temporal resolution. This, however, limits the applicability of these results to systems with small particle-fluid density ratios. The aim of the present study is to realize, in a controlled laboratory environment, conditions comparable to those experienced by frozen hydrometeors falling in the atmospheric surface layer. Using a turbulence chamber forced by hundreds of individually actuated jets, we create a large region of homogeneous air turbulence with negligible mean flow. We consider 1 mm disks to approximate plate snow crystals, which we drop in the chamber through a 3-m chute. High-speed laser imaging is performed with two cameras to obtain time-resolved images with different window sizes and resolutions. This allows us to capture the carrier fluid flow from the integral scales down to the Kolmogorov scales using particle image velocimetry (PIV). Simultaneously, particle tracking velocimetry (PTV) is used to determine the instantaneous locations of the disks and their orientation. We investigate the turbulence effect on the velocity, acceleration, rotation rate, and fall pattern of the disks, as well as their backreaction on the flow.   

Sedimentation of spheroidal particles in density stratified fluid

The effect of fluid density stratification on the sedimentation of a rigid object is more complex compared to its homogeneous counterpart. Previous studies have provided a better understanding of the physics behind the settling of a spherical body, but we know little about the effect of object shape anisotropy on its sedimentation in a stratified fluid. To this end, we consider the sedimentation of spheroidal isolated particle in a viscous density stratified fluid via numerical simulations using the Immersed Boundary Method. In a homogeneous fluid, the spheroidal particle attains a terminal velocity with a broadside on (broadside horizontal) orientation. However, the fluid stratification leads to velocity deceleration with broadside on orientation until a threshold in the settling velocity, at which the particle changes its orientation from broadside on to edgewise (broadside vertical). In addition, stratification leads to the elimination of the path instabilities which are otherwise present for spheroidal particles settling in a homogeneous fluid for the same Reynolds number. We analyze the flow and density fields to explain these results. These findings can help in understanding the settling of marine snow or anisotropic phytoplankton in oceanic environments.   

Alignment statistics of rods with the Lagrangian stretching direction in a turbulent channel flow

We study the alignment of slender rods with the Lagrangian stretching direction in a turbulent channel flow. Our objective is to understand how the distribution of relative angles between a rod and the Lagrangian stretching direction depends on the aspect ratio of the rod and upon the distance of the rod from the channel wall. We find that this distribution exhibits a plateau at small relative angles, and power-law tails at large relative angles. A plateau in the distribution of the relative angle corresponds to random uncorrelated motion. Power-law tails in the distribution result from large excursions. Slender rods near the channel centre are found to align better with the Lagrangian stretching direction compared with those near the channel wall. These observations are explained in terms of simple statistical models based on Jeffery's equation. Near the channel centre we use a two-dimensional toy model which explains the qualitative mechanisms at play. By contrast,near the channel wall, we find quantitative agreement between the statistical model calculations and simulations.   

Dynamics of straight and curvy long fibers in turbulent channel flow

In this study, we investigate the behavior of long fibers in turbulent channel flow. The experimental facility consists of a closed water channel (aspect ratio 10) and the experiments are performed at the shear Reynolds number $Re_\tau=350$. The fibers consist of neutrally buoyant rods, both straight and curvy, characterized by a length-to-diameter ratio of 120. Fibers are recorded by four high-speed cameras in a fully developed flow section. We propose a new three-dimensional reconstruction and tracking method based on the light intensity distribution, which is obtained with the multiple algebraic reconstruction technique (MART). The fibers are first discriminated, i.e. the voxels corresponding to the location of the fibers are detected, then the orientation of the fibers is identified and tracked. We show that the tomographic analysis proposed allows an accurate reconstruction of both position and orientation of the fibers. We investigate the behavior of the fibers in two distinct regions of the flow (the channel center and the near-wall region). We analyze the effect of the curvature of the fibers on their orientation and rotation rate, and we observed that in both cases the impact of the curvature is remarkable.   

Mass transfer to freely suspended particles

When small, rigid particles are suspended in a fluid, they slip, tumble and spin relative to the fluid. Simultaneously, material may be transferred from the surface by convection and diffusion. Practical examples include nutrient uptake by planktonic osmotrophs and the growth of crystals in agitated suspension. We present a generalised, theoretical approach to determine the mass transfer rate from small, ellipsoidal particles at large Peclet numbers. The approach accounts for preferential alignment between the particle and the surrounding flow field as well as its geometry. We complement these predictions with numerical simulations of scalar transport from small particles embedded in steady and turbulent flows. The results show that shape plays a significant role in the optimisation of transport from the surface of the particle. Our findings pave the way for simplified numerical models and experimental measurements of scalar transport from small particles with complex geometries.   

Accelerated Settling of Swimming Plankton in Turbulence

The chain formation of some harmful plankton species is suspected as a self-adaptive behavior to the turbulent environment, which increases their motility and enhance the chance of survival in harsh turbulent environment. However, the hydrodynamic mechanism behind this phenomenon is still unclear. For the first time, we consider the coupling of settling and swimming effect on swimming plankton in turbulence, and find the swimming accelerates the settling velocity of prolate plankton. We model plankton as prolate spheroidal swimmers, and we analysis their motions in homogeneous isotropic turbulent flow with direct numerical simulation. Settling and swimming plankton preferentially sample downwelling regions, and tend to swim in the direction of gravity, which accelerates the settling. The accelerated settling occurs when the swimming velocity is comparable to the Kolmogorov velocity scale of turbulence. Plankton swim faster when forming chains, which accelerates the settling to avoid the unbeneficial regions with intensive turbulence. On the other hand, plankton can break up and shorten the chains to reduce settling and stay in the regions suitable for their growth and reproduction.   

Extreme anisotropy: how does a sheet-like particle of atomic thickness behave in a shear flow?

Graphene and other 2D nanomaterial interact with liquids in applications ranging from graphene inks to nanocomposites, forming suspensions of sheet-like particles of extreme aspect ratios. We considered the rotational dynamics of nanometrically thin nanoplatelets in simple shear flow, simulated via Molecular Dynamics and Boundary Integral simulations. When the surface of the platelet exhibits slip lengths$\lambda $is larger than the particle thickness, the classical tumbling dynamics predicted by Jeffery is suppressed: the particle attains a stable orientation without rotating [Kamal, Gravelle, Botto. Nat. Comm. 11.1, 2020]. This effect is robust for 2D nanomaterials - graphene has $\lambda =O(10nm)$- and can leads to a substantial reduction in the effective suspension viscosity. We have identified 3 regimes of orientation depending on the rotational Peclet number Pe. Although a stable orientation is achieved only for Pe \textgreater \textgreater 1, slip affects the orientational distribution function for a wide range of intermediate Pe numbers, by decreasing the probability of observing complete rotation events. The talk will also discuss why graphene research is a fertile area for the multiphase flow community to make technological impact.   

Lord Kelvin's Isotropic Helicoid

Nearly 150 years ago, Lord Kelvin proposed the isotropic helicoid with isotropic yet chiral interactions with a fluid so that translation couples to rotation. A 3D-printed implementation of his design is found experimentally to have no detectable translation-rotation coupling, although the particle point-group symmetry allows this coupling. These results are explained by demonstrating that in Stokes flow, the chiral coupling of such isotropic helicoids made out of non-chiral vanes is due only to hydrodynamic interactions between these vanes and therefore is small. Kelvin's predicted isotropic helicoid exists, but only as a weak breaking of a symmetry of non-interacting vanes in Stokes flow.   

3D reconstruction of flat, anisotropic particles in turbulence

The past few years of research have seen a significant advancement in the understanding of inertial particles in turbulence. Understanding particle motion in turbulent flow is important for both industrial and environmental applications such as turbulent drag-reduction and the locomotion of planktonic organisms in the ocean. Research in this area has focused mainly on fibers, with some groups looking into the motion of low aspect ratio particles. Here, we investigate the motion of flat, non-axisymmetric particles. We present a method for the 3D reconstruction of flat particles. Using this method, we are able to recreate the particle orientation over time, which gives us insight into the evolution of a particle's tumbling and spinning and the influence of the particle shape on its rotation.   

Settling of inertial non-spherical particles in wavy flow

We experimentally investigated the settling of plastic rods, disks, and spheres in wavy flows. We find that the vertical velocities of the particles are dependent on particle inertia and particle shape. We characterize particle inertia with a particle Reynolds number \textit{Re}$_{p}$ defined using the particle length and the vertical velocity of the local flow. Two factors that potentially contribute to the behavior of particle velocities are the relative velocities between particles and the flow and the manner in which particles sample the flow. We find that the average relative velocities between the particles and the flow remain constant with \textit{Re}$_{p}$, even though the relative velocities of the rods vary more with orientation as \textit{Re}$_{p}$ increases. The variation of the particle velocities with \textit{Re}$_{p}$ can be explained instead by how the particles sample the flow, as each of the particle shapes nonuniformly sample the flow as a function of \textit{Re}$_{p}$. Accounting for the variation of particle settling velocities with shape and inertia is necessary to improve the accuracy of model predictions of the transport of microplastics in the ocean.\\