Bulletin of the American Physical Society
2005 58th Annual Meeting of the Division of Fluid Dynamics
Sunday–Tuesday, November 20–22, 2005; Chicago, IL
Session NE: Multiphase Flows: General IV |
Hide Abstracts |
Chair: Tristan Burton, United States Naval Academy Room: Hilton Chicago Continental B |
Tuesday, November 22, 2005 11:01AM - 11:14AM |
NE.00001: Single polymer in fluid flows: a numerical study Alberto Puliafito, Antonio Celani, Konstantin Turitsyn Thanks to the improvements in experimental techniques, a comprehensive study ofthe motion of single molecules in fluid flows has recently become a reality. The direct comparison of experiments with numerical simulations and theory permits to check and refine the various polymer models, and unveils the role of interactions between different macromolecules and between the molecule and the surrounding fluid. The dynamics of single molecule in fluid flows has been studied in great detail by Chu, Larson, Shaqfeh and coworkers. We show how nontrivial aspects of polymer-fluids interactions may be accounted for and even explained at a semi-quantitative level by means of simple, few degrees of freedom models. We investigate the dynamics of orientation and elongation of a polymer molecule in a laminar shear flow as a function of the flow strength, and we analyze the aperiodic rotational motion of the molecule. In the laminar elongational flow and in the random chaotic flow we investigate the transition time from a given initial condition to the stationary distribution as a function of the flow strength, temperature, and length of the polymer. We give a theoretical explanation for the anomalous increase of relaxation times around the coil-stretch transition. [Preview Abstract] |
Tuesday, November 22, 2005 11:14AM - 11:27AM |
NE.00002: WITHDRAWN: Simulation of Particle Interactions in DNA-laden Flows Michael Bybee, Greg Miller, David Trebotich We present a numerical method to simulate dynamics of DNA in array microchannels. Discrete DNA molecules are represented by a bead-rod polymer model and are fully coupled to the fluid. The fluid model is based on a high resolution finite difference method on Cartesian embedded boundary grids. We introduce a screened Coulomb potential between rods to model particle-particle and particle-structure electrostatic interactions while also accounting for excluded volume effects. We justify the use of this method by performing a statistical comparison of this ``soft'' potential approach to a previous ``hard'' constraint approach where particle-structure interactions were treated by elastic collision. This work was performed under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48. [Preview Abstract] |
Tuesday, November 22, 2005 11:27AM - 11:40AM |
NE.00003: Hydrodynamic Penetration of Viscous Fluids Thomas Ward, Howard Stone Using boundary layer theory we develop a model for liquid penetration driven by a source above a fluid interface and locally (along the interface) the velocity field is pure extensional flow characterized by a stagnation point. The two fluid boundary layer analysis yields data for the interfacial stress as a function of the absolute and kinematic viscosities. For systems with finite surface tension we solve the normal stress condition in the limit of small deformations and we present data for the interface shape as a functions of the Weber and Reynolds numbers. [Preview Abstract] |
Tuesday, November 22, 2005 11:40AM - 11:53AM |
NE.00004: A multi-scale study of moving contact-lines Xiaobo Nie, Mark Robbins, Shiyi Chen A continuum-atomistic multi-scale method has been developed to study moving contact-lines based on our previous hybrid method (X. B. Nie et al, JFM 2004 V. 500, pp 55-64, Phys. Fluids 2004, V. 16, pp 3579-3591.). A small inner region near the contact-line is simulated by molecular dynamics and the large remaining region is described using the Navier-Stokes equations with no-slip boundary conditions. Results are presented for two immiscible fluids confined in a channel between two rigid substrates. The dependence of contact angle on capillary number, channel width, and distance from contact line have been studied and compared with analytic and numerical solutions of the Navier-Stokes equations supplemented with heuristic models. [Preview Abstract] |
Tuesday, November 22, 2005 11:53AM - 12:06PM |
NE.00005: A unified expression for the dependence of single- and multiphase permeability on capillary pressure Bojan Markicevic, Ned Djilali, Zhong-Sheng Liu It has been shown experimentally (Katz and Thompson, Phys. Lett. B, 34:8179,1986) that the single-phase permeability of a porous medium is proportional to the square of the characteristic length devised from the capillary pressure curve. We extend this approach for displacement type flows to predict single-phase and mobile phase permeabilities simultaneously, by using the capillary pressure at the breakthrough point to calculate: (i) characteristic length, and (ii) equilibrium saturation. These two parameters are combined and a novel set of mixing rules is formulated. This framework is tested using a set of numerical results obtained from capillary network simulations using a model that accounts for invasion percolation with trapping in a drainage flow. These simulations show that with increasing medium heterogeneity, the porous medium permeability (single-phase) decreases, whereas the mobile phase permeability increases. The size of the domain is also found to influence the equilibrium saturation. Numerical results obtained for a range of domain sizes and heterogeneities are compared with predictions from the analytical model and excellent agreement is found. Based on the single expression unifying single and multi-phase flow, we are able to show that the multiphase flow is similar in nature to a single-phase flow occurring at the distinct scales. [Preview Abstract] |
Tuesday, November 22, 2005 12:06PM - 12:19PM |
NE.00006: Immunomagnetic separation in microchannels -- from MEMS to bioNEMS. Ashok Sinha, Ishwar Puri, Ranjan Ganguly Operation of a ``proof of concept'' microfluidic sensor for detecting water-borne pathogenic agents is tested. The sensor is based on a scaled down version of `\textit{immunomagnetic separation}'. The fundamental operation of the sensor consists of two miniaturized processes -- mixing of the microspheres with the sample fluid and their separation from the background fluid. The microspheres are allowed to mix with the sample fluid such that the pathogens interact with the coated microspheres. Subsequently, the pathogen-microsphere conjugates are `separated' from the background water sample by strategically placed micron sized electromagnetic traps on the channel walls. The conjugates may be transported to other sections of the sensor for detection and analysis. Sensitivity of the MEMS based sensor will depend on the extent of the initial mixing in the mixer and subsequent particle separation in the separator. While mixing in microchannels has drawn much research attention over the years, magnetic separations at this scale is not well characterized. A viable microfluidic network is conceptualized based on flow requirements for efficient magnetic particle entrapment. The use of magnetic microspheres was observed to enhance cross stream mixing in the otherwise laminar microchannel flow [Preview Abstract] |
Tuesday, November 22, 2005 12:19PM - 12:32PM |
NE.00007: Irreversibility of shear flow carrying solid particles Peter Vorobieff, Marc Ingber, Andrea Mammoli, Todd McCollam We present an experimental and numerical study of a low-Reynolds number flow in a wide-gap Couette device. Five spheres are placed into the device in a tightly packed arrangement. The inner cylinder is rotated five revolutions counterclockwise and subsequently five revolutions clockwise to assess the irreversibilities introduced into the system by particle interaction. The experimental results are compared to numerical simulations performed with a lubrication-correctly completed double layer boundary element method with a particle roughness model. The quantitative measure of irreversibility we use for comparison is based on the radial spread of the particles. If the particle roughness in simulations is adjusted to best match the experiments, a good agreement is achieved. However, the roughness value producing the best agreement is appreciably different from the measured average roughness value of the particles used in experiment. [Preview Abstract] |
Tuesday, November 22, 2005 12:32PM - 12:45PM |
NE.00008: Chaining of Dielectrophoretic Suspensions under Simple Shear Howard Hu, T.N. Swaminathan Polarizable particles in suspensions are known to form chains when subjected to an electric field. A finite element based numerical method to evaluate the dielectrophoretic forces between particles, using the Maxwell's Stress Tensor method, has been implemented. This method is known to provide more accurate dielectrophoretic forces between the particles when compared with the traditional dipole moment approximations, especially when the interparticle gaps are small. The difference of the predicted dielectrophoretic force between two particles from our numerical solution and from the dipole moment approximation is observed to increase as the gap between them decreases. Numerical simulations demonstrate the formation of chains in a dielectrophoretic suspension under the action of an electric field. When this suspension is under simple shear, the chain tends to deform and eventually break. The final length of the chains is a function of the Mason number, which is the ratio of the viscous forces to the electrical forces experienced by the particles in the suspension. The dependence of this length on the Mason number has been determined from numerical simulations and compared with traditional analytical chain models. [Preview Abstract] |
Tuesday, November 22, 2005 12:45PM - 12:58PM |
NE.00009: Surface drag in a fluidized bed Daniel Goldman, Wyatt Korff Animal locomotion on sand involves drag at the sand surface for a range of substrate conditions. Inspired by this, we study drag of a half-submerged 2 cm disk using a large aspect ratio (1200x800 particle diameters) air fluidized bed of $250 \mu m$ glass beads to control the properties of the granular material. We vary the air flow rate $Q$ to the bed and the drag velocity $v_d$ (0-40 cm/sec) of the disk. Below fluidization, the drag force $F_d$ increases linearly with $v_d$, with nonzero intercept; the intercept decreases as fluidization onset is approached. Above onset, $F_d$ is no longer linear in velocity, but has positive curvature. For large enough $v_d$, we observe the formation of a wake behind the disk. We find a sharp onset in drag associated with this wake after removing the viscous drag, similar to studies of wave drag in a viscous Newtonian fluid\footnote{T. Burghelea and V. Steinberg, Phys. Rev. Lett. {\bf 86}, 2557, (2001)}. The existence of an onset to wake formation resulting in rapid increase in drag in fluids at a critical velocity is a result of the competition between surface tension and gravitational restoring force; in the fluidized cohesionless grains it is not clear what mimics the effect of the attractive force. [Preview Abstract] |
Tuesday, November 22, 2005 12:58PM - 1:11PM |
NE.00010: Forces on a Spherical Particle in Linear Shear Flow Inchul Kim Accurate prediction of particle dispersion is important in many two-phase flows. When particles are submerged in a shear flow, there are lateral forces on the particles, and these lateral forces affect the dispersion of the particles very much. Saffman (1965) derived an expression for the lift force on a spherical particle for Re$<<$1. The lift force in Saffman's expression is toward the higher velocity side in a shear flow. Subsequently, Dandy and Dwyer (1990), Kurose and Komori (1999), and Bagchi and Balachandar (2002) performed numerical investigation for finite-Reynolds-number flows. All these authors used a small computational domain ranging 20 to 30 sphere radii. The result by Dandy and Dwyer shows that the lift force is toward the higher velocity side. The results by Kurose and Komori and by Bagchi and Balachandar show that the sign of the lift force changes, respectively, at Reynolds number about 14 and 55.5 for the dimensionless shear rate 0.1. To provide correct numerical data for the forces on a spherical particle in linear shear flow, the present author has performed an accurate numerical computation for$1\le Re\le 200$. The effects of computational domain size and space resolution were examined, and accurate converged numerical results have been obtained. [Preview Abstract] |
Tuesday, November 22, 2005 1:11PM - 1:24PM |
NE.00011: Capillarity and capillary waves in immiscible two-fluid channel flows Philip Yecko The configuration of two immiscible fluids undergoing sheared
Couette-Poiseuille
flow in a straight channel is analyzed as a linearized stability
problem, both from
the point of view of eigenvalues and energy growth resulting from
the non-normality
of the stability operator. Only mild density contrasts
$(0.75 |
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