Bulletin of the American Physical Society
2006 59th Annual Meeting of the APS Division of Fluid Dynamics
Sunday–Tuesday, November 19–21, 2006; Tampa Bay, Florida
Session AE: Multiphase and Particle-Laden Flows I |
Hide Abstracts |
Chair: James Riley, University of Washington Room: Tampa Marriott Waterside Hotel and Marina Florida Salon 123 |
Sunday, November 19, 2006 8:00AM - 8:13AM |
AE.00001: Turbulence attenuation by high inertia particles having the same response time Takuya Tsuji, Toshitsugu Tanaka Interactions between particles and turbulence are very complicated and physical mechanism of turbulence modulation by particles is still not known well. In this study, turbulence attenuation by high inertia particles that have the same response time is investigated in stationary isotropic homogeneous turbulence. An immersed-boundary DNS technique which resolves the flow around each particle properly was utilized. Particles were set to slightly larger than Kolmogorov micro scale. All cases show turbulence attenuation while the magnitude of turbulence decay varies on particle Reynolds number. When particle is relatively large, vortex tubes exist as evading particles. On the contrary, when particle becomes small, growth of vortex tubes are blocked. This leads higher attenuations. [Preview Abstract] |
Sunday, November 19, 2006 8:13AM - 8:26AM |
AE.00002: Interaction of a finite-sized stationary particle with turbulent channel flow Lanying Zeng, S. Balachandar, Fady Najjar, Paul Fischer The interaction of a {\em {finite-sized}} particle with a turbulent channel flow is considered using DNS. Different particle sizes and particle locations in the channel are used to study the interaction. The particle Reynolds number is varied from 40 to about 450, and the particle is fixed either in buffer region or at the channel center. A spectral element methodology is used to discretize the channel exterior of the embedded sphere. Great care is taken to ensure complete resolution of all the relevant length scales of the wall turbulence, attached boundary layers and the unsteady wake behind the sphere. In this talk, we will present the effect of the turbulent channel flow on the particle wake and vortex shedding, and also the modulation of turbulent channel flow in the particle wake. The results will be compared and contrasted with the similar study for isotropic turbulence. [Preview Abstract] |
Sunday, November 19, 2006 8:26AM - 8:39AM |
AE.00003: Experimental study on the interaction between Burgers vortex and a solid particle using 2D PIV measurement Yohsuke Tanaka, Takuya Tsuji, Toshihiro Kawaguchi, Toshitsugu Tanaka, Yutaka Tsuji Many researchers mainly study on the statistical properties of the turbulence and kinematics of particle motion. In the present study, we focus on the elementary step of an interaction between single Burgers vortex tube and a heavy particle. Particular attention will be paid to understanding how heavy particle influences Burgers vortex, and how the vortex is strengthened. Burgers vortex is the simplest model of tube-like coherent structures. Burgers vortex represents one of the few known exact solutions to the full Navier-Stokes equations, and the vortex have two-dimensional vorticity field. This vorticity change is observed by 2D PIV measurement. There are many situations where the particle size dp is the order of Kolmogorov scale or larger, and the particle Reynolds number Rep is larger than unity. The effect of finite particle size, particle wake, and vortex shedding are important in this range. We carried experiments on conditions of 150 $<$ Rep $<$ 1400 and 0.5 $<$ dp/$\eta <$ 1.5, and we observed the influence of the distance between vortex and particle. [Preview Abstract] |
Sunday, November 19, 2006 8:39AM - 8:52AM |
AE.00004: DNS of Multiphase Forced Isotropic Turbulence Lin Zhang, S. Balachandar, Paul Fischer Particle turbulence interaction is of fundamental importance. However, theoretical and computational studies have been generally constrained to dilute dispersion of very small particles. Our understanding of this problem in the regime where the particles are of {\em{finite-size}}, has been quite limited. In particular, influence of turbulence on the particles, back effect of particles, and inter particle effect within a distribution, are all open questions in the context of finite-sized particles. A higher-order accurate Spectral-Element-Methodology (SEM) is used in the {\em{fully resolved}} simulations of forced isotropic turbulence. 100 randomly distributed spheres, of the Taylor microscale size, are embedded in the computational domain. We had developed an efficient technique to automatically discretize the domain with randomly distributed spheres into body-fitted hexahedral elements. The subelement resolution to chosen to fully resolve all the turbulent scales, attached boundary layers on the sphere and their wakes. %The meshing algorithm is improved to generate grids for %truly randomly distributed spheres. %A resolution study is carried on %by simulating the single phase isotropic turbulence %using both SEM and a fully spectral code. %The finest resolvable scale by the spectral code %is also resolved by SEM at a fine discretization. Employing the same random forcing, we perform the isotropic turbulence simulation in a ${(2\pi)}^3$ box both with and without the randomly distributed spheres. Through comparison of the energy spectra, two-point correlations, force statistics, etc, particle turbulence interactions is explored. [Preview Abstract] |
Sunday, November 19, 2006 8:52AM - 9:05AM |
AE.00005: The effects of upstream turbulence on flow through random arrangements of spheres Ying Xu, Madhusudan G. Pai, Shankar Subramaniam We perform Direct Numerical Simulations (DNS) of flow at moderate Reynolds numbers (based on mean flow velocity and particle diameter) through random arrangements of stationary spheres with varying levels of upstream turbulence. A pseudo-spectral implementation of the immersed boundary method is used to solve the Navier-Stokes equations with exact boundary conditions on each particle's surface. The simulations are used to probe the effects of upstream turbulence on the mean drag force experienced by the particles. Similar calculations for flow past a single sphere by Yusof (PhD Thesis, Cornell Univ. 1996) show that upstream turbulence can destabilize the wake behind the sphere and result in increased the drag. These calculations allow us to study the combined effects of particle shielding and upstream turbulence. [Preview Abstract] |
Sunday, November 19, 2006 9:05AM - 9:18AM |
AE.00006: A numerical study of subgrid-scale effects on solid particle motion and heat transfer in a dilute, particle-laden turbulent flow. Saensuk Wetchagarun, James Riley The effects of carrier-phase, subgrid-scale (SGS) velocity and temperature on particle motion and heat transfer are investigated via \textit{a priori} testing using direct numerical simulation (DNS). The carrier phase obeys the incompressible form of the Navier-Stokes equations. The flow is assumed to be dilute so that one-way coupling is implemented. The filtered carrier phase data for decaying, isotropic turbulence are obtained by filtering the output from the DNS. Particles are individually tracked, and their temperature and heat transfer with the local fluid is computed by solving an energy equation. Both the large-scale and the SGS fluid properties as `seen' by the particle are computed directly from simulation results. It is found that, depending on the size of the filter and the local value of the Stokes number, the SGS motions can significantly affect the particle motion and the heat transfer. In addition, the SGS effects depend strongly on the Stokes number, something not included in some recent models. Improvements in the SGS modeling, based upon these numerical results, are suggested. [Preview Abstract] |
Sunday, November 19, 2006 9:18AM - 9:31AM |
AE.00007: Inertial particle clustering in turbulence: A comparison of experimental results with theory E.W. Saw, R.A. Shaw, S. Ayyalasomayajula, Z. Warhaft In order to test the validity of recent theoretical and computational (DNS) predictions on inertial particle clustering in turbulence, we have investigated the spatial distribution of particles (water droplets) in high-$R_\lambda$ laboratory turbulence. The experimental facility is an active-grid wind tunnel, generating nearly homogeneous and isotropic turbulence with maximum $R_{\lambda}\approx 10^3$. Droplets are sprayed into the flow and downsteam their diameter, longitudinal speed, and arrival time are recorded with a phase Doppler interferometer. We calculate pair correlation functions $\eta(r)$ conditioned on droplet diameter $d$, to show how the scale-dependence of clustering changes with Stokes number ($St= (1/18)(\rho_d/\rho)(d/r_k)^2$, where $r_k$ is the Kolmogorov microscale, and $\rho_d$ and $\rho$ are the droplet and air mass densities, respectively). We then compare the scale dependence of the pair correlation functions with various power law relations suggested from previous theoretical work, all of which have the form $\eta(r) \propto (r/r_{k})^{-f(St)}$. [Preview Abstract] |
Sunday, November 19, 2006 9:31AM - 9:44AM |
AE.00008: Temporal Evolution of Particle Clustering in Isotropic Turbulence Lujie Cao, Jeremy de Jong, Juan Salazar, Lance Collins, Scott Woodward, Hui Meng The particle radial distribution function (RDF) has been identified as a key variable for quantifying the effect of clustering on binary processes such as collision.~Measurements of the RDF are best done in three dimensions (Holtzer {\&} Collins 2002), hence most results to date come from direct numerical simulations at modest Reynolds numbers. To complement our DNS database, we apply digital holographic particle image velocimetry (DHPIV) to perform three-dimensional measurements of particle clustering in nearly homogeneous isotropic turbulence in a box.~The turbulence corresponding to different fan speeds has been characterized using particle image velocimetry (PIV), and the turbulent energy dissipation rate was obtained from a fit of the second-order structure function. Metal-coated hollow glass spheres with a well defined particle size distribution were injected and three-dimensional DHPIV snapshots were obtained and analyzed.~The three-dimensional RDF was observed to increase with time from injection.~By phase averaging the measurements based on the time from injection, it was possible to semi-quantitatively measure the temporal evolution of the RDF.~This is believed to be an important feature of the RDF in practical industrial applications (e.g., powder manufacturing) and naturally occurring flows (e.g., cloud droplets), where temporal dynamics may result from changes in the local conditions and/or droplet coalescence (Reade {\&} Collins 2000). [Preview Abstract] |
Sunday, November 19, 2006 9:44AM - 9:57AM |
AE.00009: Comparison of Direct Numerical Simulation and Experimental Observation of Particle Clustering in Isotropic Turbulence Juan Salazar, Jeremy de Jong, Lujie Cao, Scott Woodward, Hui Meng, Lance Collins ~Experimental observation of particle clustering in turbulent flow is often complicated by particles of non-uniform in size, and the accuracy of most diagnostic techniques is sensitive to the particle size. To make meaningful comparisons with experiments direct numerical simulations (DNS) must take into consideration the properties of the particles and model the limitations introduced by the diagnostics.~We present a series of DNS of inertial particles in turbulence that were designed to match the measurements of Cao et al.~In their experiments, the particle radial distribution function (RDF) was obtained from 3-D holographic images of particles in a turbulence box.~DNS was done using the same particle size distribution as the particles in the experiments. The parameters of the flow (Re {\#}) and the particles (St {\#}) were matched.~ In the DNS, we eliminated particles below a specified size and recalculated the RDF to better match the experiments.~Trends in the variation of the RDF with size cutoff were found to be non-intuitive,~but can be explained based on an extension of the theory of Chun et al. (2005) to polydisperse particles.~We optimized the cutoff size based on the comparison of the experimental and numerical RDF at one fan speed and applied the same cutoff to other fan speeds. The results showed good agreement.~The comparison highlights the complexity of matching DNS and experimental observations. [Preview Abstract] |
Sunday, November 19, 2006 9:57AM - 10:10AM |
AE.00010: Simultaneous measurements of droplet clustering and turbulence in cumulus clouds R.A. Shaw, K. Lehmann, E.W. Saw, H. Siebert Droplet clustering in clouds due to turbulence may act to accelerate the formation of rain through droplet coalescence. Theory and computations suggest that clustering strength depends on the droplet Stokes number, which is proportional to $d^2 \varepsilon^{1/2}$, where $d$ is droplet diameter and $\varepsilon$ is turbulence energy dissipation rate. To evaluate this hypothesis we have made simultaneous (in time and space) measurements of droplet spatial distribution, droplet size distribution, and turbulent velocity. The latter allows calculation of local energy dissipation rate. Measurements were made using the Airborne Cloud-Turbulence Observation System (ACTOS) deployed via helicopter. To increase confidence in the clustering measurements two instruments, collocated in space, and based on different operating principles and measurements geometries, were used for the clustering measurements. Clustering signatures are present and tend to increase with droplet Stokes number. The role of gravitational sedimentation in modulating inertial clustering of cloud droplets is explored. [Preview Abstract] |
Sunday, November 19, 2006 10:10AM - 10:23AM |
AE.00011: Aggregate formation in 3D turbulent-like flows A. Dominguez, H.J.H. Clercx Aggregate formation is an important process in industrial and environmental turbulent flows. In oceans turbulence play an important role on Marine Snow (aggregate) formation. For a proper description, the study of aggregate formation in turbulent flows requires a particle based model i.e. following trajectories of single particles. For these to be done, it is required to model three main processes: the flow, the motion of the particles and the encounter and coalescence of particles. In this study we use 3-D kinematic simulations to model the turbulent flow. A simplified version of the Maxey-Riley equation is used, with Stokes drag, bouyancy and added mass forces. In the collision and aggregate formation module a geometrical collision check is used. When the distance between two particles, is smaller than the sum of their radii, a collision takes place. All the particles that collide stay together to form an aggregate. To account for the porosity of the aggregates a Fractal Growth Model is used. In this study we will explore the effects of different parameters on the aggregate formation (\textit{e.g. St; Wst; $\phi $; Re}) and the effects of two different background populations: constant and decaying. [Preview Abstract] |
Sunday, November 19, 2006 10:23AM - 10:36AM |
AE.00012: Relative acceleration of particle pairs in flows past random arrays of spheres Madhusudan G. Pai, Rahul Garg, Shankar Subramaniam Hydrodynamically-induced particle clustering is observed in inertial suspensions of particle-laden flows. Second-order statistics like the pair correlation function are used to quantify the effect of clustering. The evolution equation for the second-order density contains a transport term involving the relative acceleration of a particle pair. In this study, we perform direct numerical simulations of flow over stationary spheres and study the effect of particle spacing on the relative acceleration of a particle pair. Understanding effects of acceleration contributions from neighboring particles is a first step towards identifying the mechanisms that contribute to the clustering phenomenon in such systems. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700