Session MP: Multiphase Flows VI

Investigating Cavitation Inception in a Flow Over an Open Cavity Using Large-Eddy Simulation

A large-eddy simulation of flow over an open cavity corresponding to the experimental setup of Liu and Katz (Phy. Fluids, {\bf 20}, 2008) has been performed. The flow Reynolds number based on the cavity length and the free stream velocity is 190,500. The single-phase, finite-volume flow solver uses five million grid points with around two million points clustered to resolve the shear layer. The flow statistics, including mean and rms velocity fields and pressure coefficients, are compared with the experimental data to show reasonable agreements. Cavitation inception is predicted using a discrete bubble and a transport equation model. The cavitation inception occurs near the trailing edge similar to that observed in the experiments. Details of the simulation and the predictive capability of the models will be discussed in detail.   

Suppression of Tip Vortex Cavitation by Water and Polymer Injection

Tip vortex cavitation (TVC) is typically the first form of cavitation observed in propellers; therefore a delay of its inception is desirable. In this study, TVC inception was delayed via mass injection from the tip of a modified NACA-66 elliptic plan-form hydrofoil. The injected mass used were water and Polyox WSR 301 solutions with concentrations from 10 to 500 ppm. The free-stream nuclei content was measured using a Cavitation Susceptibility Meter, and the TVC inception event rate was quantified with a light scattering system. The vortical flow field in the region of inception was characterized using Stereo Particle Imaging and Velocimetry (SPIV), with measurements taken from 0.25 to 1 chord length at various concentrations and injection rates. It was observed that TVC inception was delayed with injection by a $\Delta \sigma $ of 0.03 to 1.8 from a baseline of $\sigma $ = 3.5. Injection with higher polymer concentrations and higher volume flux led to larger delays. A saturation effect for the TVC suppression was observed for both the polymer concentration (125ppm) and volume flux rate ($\dot {Q}_j $\textbf{/}$\dot {Q}_c $ = 0.48). These effects are related to the SPIV flow field.   

Effects of surface characteristics on hydrofoil cavitation

This was an exploratory research project aimed at capitalizing on our recent research experience with unsteady partially cavitating flows. Earlier work identified the significant and unexpected effect of surface properties and water quality on the dynamics of these flows. The aim of this study was to explore the possibility of using hydrophobic surfaces to control or minimize unwanted vibration and unstable operation in the partially cavitating regime. A candidate shape, denoted as the Cav 2003 hydrofoil was selected on the basis of theoretical analysis for a given range of contact angle. We manufactured three hydrofoils of identical cross section, but different surface characteristics. Three different surfaces were studied: anodized aluminium (hydrophilic), Teflon (hydrophobic), and highly polished stainless steel (hydrophobic). Contact angle was measured with a photographic technique developed by three of the undergraduates working in the project. Studies were made in both weak and strong water. Significant surface effects were found, but were unexpected in the sense that they did not correlate with measured contact angles.   

Investigation of the behavior of a ventilated supercavity

The topic of supercavitation is of considerable interest to drag reduction and/or speed augmentation in marine vehicles. During the present experimental work, the ventilated supercavity formed behind a sharp-edged disk is investigated utilizing several different configurations. Results regarding cavity shape, cavity closure and ventilation requirements versus cavitation number and Froude number will be presented. Additionally, effects related to flow choking in a water tunnel test section are discussed. Results obtained are similar in character to previously reported results, but differ significantly in measured values. Cavity shape, particularly aft of the maximum cavity diameter, is found to be a strong function of the model support scheme chosen.   

Contact model for a normal immersed collision between a solid sphere and a wall

Particle collisions in a liquid environment play an important role in the study of liquid-solid flows. This talk focuses on the collision process and presents simulations of the coupled motion of a solid particle and surrounding liquid. The simulations use an immersed boundary method with axisymmetric coordinates to investigate the normal collision between a rigid sphere and a wall in viscous liquid. A contact model including the elasticity during the solid contact is introduced into the simulation when the distance between the sphere and the wall falls below a certain value, which enables the calculation to reproduce the rebound process with calibrated model parameters. Simulation results are compared with experimental measurements of the impact and rebound trajectories for a range of Stokes numbers. The long term goal of this work is to develop collisional strategies that can be incorporated into simulations involving many particles in a viscous fluid.   

Can the Cahn-Hilliard model quantitatively describe moving contact lines?

Diffuse-interface models may be used to compute moving contact lines because Cahn-Hilliard diffusion regularizes the singularity at the contact line. This paper investigates the basic questions underlying this approach. Through dimensional analysis and numerical computations, we demonstrate that the Cahn-Hilliard model approaches a sharp-interface limit when the interfacial thickness is reduced below a threshold while other parameters are kept fixed. In this limit, the contact line has a diffusion length that is related to the slip length in sharp interface models, and the relaxation of wall energy determines the deviation of the dynamic contact angle from the static one. From these results, we develop practical guidelines for attaining the sharp-interface limit in numerical simulations and for quantitatively reproducing experimental data on the apparent contact angle. With the wall relaxation parameter fitted to a single experimental measurement, the model gives an excellent representation of experimental data over the entire range of flow rates up to wetting failure.   

Role of ions in thermal diffusion of DNA: Lattice Boltzmann based simulations

The Ludwig-Soret effect, the migrarion of a species as a consequence of a temperature gradient, has been a factor in the development of microfluidic laboratory instrumentation. In a system consisting of DNA in a buffered salt solution exposed to a temperature gradient in micro channels, it has previously been observed that DNA will migrate to the colder regions, yielding an irregular density profile. We present a computational model in order to quantify the motion of the particles and describe the causes of this migration. In this construct, the salt ions are modeled as charged point particles and DNA as charged beads connected by springs. The motions of particles is calculated by using a combination of Brownian dynamics and the lattice Boltzmann method. We observe that the salt are also affected by the temperature gradient, creating a density profile. By varying the number of ions, the charge of the ions and the length of the DNA chain, we observe that the accumulation of ions in the cold region enhances the migration of the DNA to those regions of the channel.   

Laser-induced liquid flow and phase change phenomena in thin metal films

We develop a mathematical model of the liquid flow and phase change phenomena involved in the laser ablation of thin metal films deposited on glass substrates. Applications include the manufacture of micro- and nanochannels for use in various microfluidic devices. Interaction of the laser beam with the metal film is a complicated process, characterized by high temperature gradients. In this work we investigate the regime in which the temperature is high enough to cause phase explosion to occur in a small region of the metal film surrounded by a pool of molten metal. To model the liquid flow in the melt region, we include the effects of both evaporation from the surface and viscous flow induced by thermocapillary stresses. Evolution of the surface of the molten film is investigated, and the impact of phase explosion on the flow is discussed.   

Lift and drag forces on a particle rolling/sliding on a wall in a shear flow at finite Re

Recent research [1] has shown that both the shear and wall-induced lift contributions on a particle sharply increase as the gap between the wall and the particle is decreased. Explicit expressions that are valid over a range of finite Re were obtained for the drag and lift forces in the limiting cases of a stationary particle in wall-bounded linear flow and of a particle translating parallel to a wall in a quiescent ambient. Here we consider the more general case of a translating particle in a wall-bounded linear shear flow where both the shear and wall effects superpose. We have considered a modest Reynolds number range of 1 to 100. Direct numerical simulations using immersed boundary method were systematically performed to figure out characteristics of hydrodynamic forces on a finite-sized moving particle whose location is nearly sitting on the wall. We present composite correlation for the drag and lift forces which are in agreement with all the available low Reynolds number theories. The effect of particle rolling or sliding on the wall is also considered and thereby the Magnus lift contribution is also considered in this work. [1] L. Zeng, Ph.D. thesis, University of Illinois at Urbana-Champaign, 2007.