We focus our attention on the low pressure area in turbulent fine vortices, which is Burgers vortex type and closely related to cavitation inception, because it has been ignored in previous methods for cavitating flow simulation. As a basis of model construction, the interaction between a Burgers vortex and cavitation is investigated by the direct numerical simulation (DNS). The vortex is artificially reproduced to represent a fine-scale elementary motion in fully developed turbulence. The results show that both the circumferential velocity and the streamwise vorticity distributions shift due to the sudden expansion by cavitation inception. Assuming the circulation is kept constant, we derive a new model which represents velocity modulation of an elementary vortex for the case of cavitation. Our model successfully explains the DNS results. This simple model can become the basic concept of cavitating turbulence model.

View full abstract

Turbulence modulation induced by bubble swarms was experimentally investigated. Homogeneous isotropic turbulence is formed by using an oscillating grid in a cylindrical acrylic vessel. The bubble size and generation frequency are completely controlled using a bubble generator. A bubble swarm is generated at a frequency of 4 Hz and consists of 19 bubbles of 2.8 mm in equivalent diameter. The liquid phase velocities were measured via two LDA (Laser Doppler Anemometer). Turbulence intensity, spatial correlation and integral scale were evaluated from the LDA data obtained by the two spatially-separate-point measurement. The turbulence intensity dramatically changed by adding bubble swarms. The original isotropic turbulence was modulated to the anisotropic turbulence by the mutual interaction between bubble swarms and ambient isotropic turbulence. The turbulence modulation was also confirmed by the integral scale.

View full abstract

The microscopic configuration of fluid interface near the wetting or dewetting front (i.e., moving contact line) is of significant interest in relation to many industrial processes. In this paper we report on molecular dynamics (MD) simulations conducted for investigation of the dependence of the microscopic contact angle θ on the speed of contact line relative to the solid surface, ΔV. Our simulations were made for a Couette flow geometry where two Lennard-Jones (LJ) liquids are sheared between two solid walls moving at a speed V in opposite directions. The two liquids are immiscible and have different wall-wetting properties. The liquid configuration remains unchanged with time, since ΔV=V at all the contact lines, for V below a certain critical value. The contact angle θ measured under such steady conditions varies monotonically with ΔV. Above the critical wall speed, however, the flow at the receding contact line, where the more wetting liquid of the two is being replaced by the less wetting one, starts becoming unsteady, as ΔV remains smaller than the critical wall speed. We show that the viscous and pressure forces measured from the MD simulation results can be reproduced by the Stokes flow model of Huh and Scriven (1971) using the MD-predicted value of θ at the given ΔV. It follows that, by the aid of Qian et al.'s formulation (2003) of the stress balance at the moving contact line, and by using the MD-predicted proportionality constant between the total wall-tangential force and ΔV, this model can reproduce the MD-predicted dependence of θ on ΔV. The model predictions, furthermore, give insight into the reason why the contact line flow becomes unsteady above the critical wall speed.

View full abstract

At a gas-liquid interface, there are many unknown physical and chemical phenomena related to thermodynamics, electromagnetics, hydrodynamics, and heat and mass transfer. Therefore, a modeling of gas-liquid interface is one of key issues of interfacial phenomena of multiphase flows. In the previous our study, having assumed that the interface is a thin membrane and has a finite thickness, we have developed a new mathematical model of the gas-liquid interface based on thermodynamics and mathematical approach. In that study, we have derived the new equation of free energy based on the lattice gas model including the influence of electric double layer on the interface caused by a contamination. Then, by using this interface model, we derive the jump condition at gas-liquid interface treated by thermodynamics. In this study, we discuss on the thermodynamic jump condition in macroscopic scale through the analytical treatment. Finally, we reveal the force vector fields on the gas-liquid interface depending on an electric potential and a distribution of the contamination at the interface.

View full abstract

Relations between drag coefficients C_D and bubble shapes are investigated by comparing our experimental results with both analytical [Moore, J. Fluid Mech., 23, 749-766 (1965)] and numerical results [Blanco & Magnaudet, Phys. Fluids, 7(6), 1265-1274(1995)] on ellipsoidal bubbles rising in a quiescent clean liquid. We clarify that the analytically obtained C_D overestimates the experimental results especially in the case of either high aspect ratio or low Reynolds number, and that numerically obtained results for ellipsoidal bubbles well predict our experimental results. We also show that the effect of the deviation in the shape from the ellipsoid, i.e., fore-aft asymmetry, on C_D is negligible; C_D for fore-aft asymmetry bubbles is in good agreement with the numerically obtained C_D for ellipsoidal bubbles.

View full abstract

A novel microbubble flotation using a spiral-liquid-flow type microbubble generator was developed to recover the iron oxide fine particles from the suspension containing dilute surfactants. Although the suspended iron oxide particles were hardly removed from suspension using other aerators such as a single orifice and a glass ball filter, the microbubble generator achieved 90% recovery of the particles in 60 minutes. In bulk liquid, both surfaces of microbubbles and iron oxide particles became electrically charged. When pH was controlled between 4.4 and 7.8, the surface of microbubbles was charged negatively and the surface of iron oxide particles was charged positively. Therefore, the microbubbles and the particles were attracted each other. At pH = ca. 5 the microbubbles adsorbed well the iron oxide particles. In the foam layer at the top of flotation column, the particles were trapped and removed from the bulk liquid. The recoverability depended on the degree of ionization and the concentration of added surfactant. In the present system, the concentration of surfactant was suitable to be only 1 to 2 % of the critical micelle concentration to achieve the highest recoverability of suspended particles from the suspension. Moreover, to understand the mechanism of the iron oxide particle separation using microbubble flotation, the novel kinetic model which considers the adsorption and the release coefficients was proposed. The microbubble flotation model could estimate well the removal of iron oxide particles from the suspension.

View full abstract

For the phosphorus recovery from the excess sludge in the sewage treatment process, the decomposition of excess sludge by using ozone micro-bubbles was focused, and the influence of ozone micro-bubble on the foam behavior and the decomposition performance was examined. A cocurrent upflow bubble column with micro-bubble generator was used and the excess activated sludge was decomposed under the various conditions. The foam behavior was observed and the decomposition performance was measured.
The foam behavior was quite different with the bubble size. When ozone micro-bubbles were dispersed, the ozone decomposition was much enhanced. As a result, the decomposition time and required amount of ozone became much lower than those in the case of milli-bubbles.

View full abstract

The electrostatic levitation technique makes it possible to measure thermo-physical properties of extremely high-temperature molten materials. In the technique, viscosity of levitated sample has been estimated from the damping constant time with an oscillating drop method based on linear approximations. However, this method is limited to low-viscosity fluids because the oscillation could not be excited to a highly viscous liquid drop. The measurement method which can be applied to viscosity over the entire range possible has not been established yet. The main purpose of the present study is to investigate the possibility to measure viscosity of entire range with a levitating liquid drop. In the present study, viscous drops were levitated by the electrostatic force. The effect of viscosity on the deformation of a rotating liquid drop was experimentally investigated. As a result, the rotational deformation of a low-viscosity drop is found to be qualitatively similar to the numerical result. On the other hand, large differences from the numerical result are observed in the case of high-viscosity drops. It is confirmed especially that viscosity is closely related to the evolution of drop shape during the drop breakup process. From the experimental results, it is clarified that the viscosity of a levitating liquid drop can be estimated from the breakup behavior of the drop.

View full abstract

Air-water two-phase swirling flows in a downscaled model of a steam separator for a boiling water nuclear reactor have been measured while paying attention to the effects of pick-off-ring (POR) shape on separation performance. As a result, the following conclusions are obtained. (1) In an annular swirling flow, the ratio W_s^* of the separated liquid flow rate to the total liquid flow rate increases with J_G and does not depend on J_L and POR shape so much. (2) Separation performance becomes lower and does not depend on POR shape so much when the liquid film thickness becomes larger than the size of the gap between the barrel wall and the POR wall. The gap size should be determined so as to be smaller than the maximum film thickness. (3) The POR shape strongly affects the separation performance when the liquid film thickness is smaller than the gap size.

View full abstract

Heat generation density from semiconductor devices increases with the rapid development of electronic technology. The cooling system using boiling two-phase phenomena attracts much attention because of its high heat removal potential. Experiments on the increase of CHF for flow boiling in narrow channels by improved liquid supply were conducted for the development of high-performance space cold plates. A large surface of 150mm in heated length and 30mm in width with grooves of an apex angle of 90 deg, 0.5mm in depth and 1mm in pitch was employed. A structure of narrow heated channel between parallel plates with an unheated auxiliary channel was devised and tested by using water for different combinations of gap sizes and volumetric flow rates, where inlet of the main heated channel and the outlet of auxiliary unheated channel were closed in order to prevent the flow instability observed frequently at low flow rate for parallel two channels. The data were compared with those for the two parallel channels. CHF values of 2×10⁶W/m² were obtained for the improved configuration. For gap sizes of 2mm and 5mm at high volumetric flow rate larger than 3.60l/min, the extension of dry-patches were observed at the upstream location of the main heated channel resulting burnout not at the downstream but at the upstream. The values of pressure drop for gap size of 2mm were higher than that for gap size of 5mm. When the performance of cooling system was evaluated by the pump power ignoring its variation of efficiency with volumetric flow rate, i.e., the power defined as the product of the pressure drop and the total volumetric flow rate, higher values of CHF were obtained for gap size of 5mm as far as the same pump power was concerned.

View full abstract

The objective of this work is to establish a constitution equation of the contact heat transfer between colliding particles for the Discrete Element Method (DEM) simulations. Since the thermal resistance model is available to estimate the contact heat transfer, the heat transfer between two particles (spheres) was measured by changing surface roughness. The result of this is that the contact heat transfer increased as surface roughness increased. This tendency is opposite to a general one in the thermal resistance model. This would be caused by the geometric condition of the spherical contact plane. Larger surface roughness yields larger contact area. This effect would be larger than that of increasing thermal resistance. Moreover, we tried to directly observe the surface under contact condition by using a laser microscope in order to examine the surface conditions in detail. The cross-sectional profile of contacting particles was successfully detected. The macroscopic contact area calculated by considering the surface roughness well agreed with the observed one.

View full abstract

In this paper a method to identify flow regime is reported in order to give a fundamental insight into air-magnetic fluid two-phase flow. The flow regime classification is based on the magnetic induction method of void fraction measurement. The proposed method identifies the flow regimes by waveforms of induced electromotive force. In experiment, we obtained induced electromotive force in actual two-phase flow by adopting volumetric flux of each phase as a main parameter. In order to confirm the flow regime, flow-regime visualization was carried out. The relationship between flow regime and the corresponding waveform of the induced electromotive force was investigated. As a result, the waveforms of electromotive force can be categorized, as in the case of flow regime, into waveforms corresponding to bubbly, slug, churn and annular flows. Moreover, the signal of induced electromotive force was treated by a signal processing with wavelet transform to obtain a probability density function (PDF).

View full abstract

An estimation of void fraction in tight-lattice rod bundles was carried out. Five types of void fraction experiments with 7-, 14-, 19- and 37-rod and rod-gap of 1.0 - 1.3 mm bundle and spacer effect tests were conducted ranging from 0.1 to 7.2 MPa. Extensibility of a system accident analysis code, TRAC-BF1 and one-dimensional drift-flux model to the tight-lattice rod bundle was studied. The TRAC-BF1 and the model calculated the void fraction with good agreement to the data in case of relatively high quality and void fraction region. Applicability of advanced numerical analysis codes, NASCA, ACE-3D, TPFIT to the tight-lattice rod bundle was verified by comparing with the three-dimensional void fraction data measured by neutron tomography. Tendency of the calculated void fraction by these codes and measured data was similar as same order of the measurement error. Vapor distribution and velocity profile of water and vapor were discussed based on data. The reason why a boiling transition phenomenon occurred at the center region of the 37-rod bundle test section is probably related to a lower local liquid holdup at the channel center region.

View full abstract

Channel stability, core-wide stability, and regional stability experiments were conducted for a wide range of operating conditions of a BWR-5 with UOX (uranium dioxide) type 9×9A fuels installed, including maximum power points along the minimum pump speed line and the natural circulation line. The SIRIUS-F facility, used in this study, was designed and constructed for highly accurate simulation of void-reactivity feedback of BWRs. The decay ratios and the resonance frequencies are in good agreement with those from the design analysis code, ODYSY. The SIRIUS-F experimental results demonstrated stability characteristics such as a stabilizing effect of the power, and revealed a sufficiently large stability margin even under hypothetical conditions of power enlargement.

View full abstract