Microbubble has distinguished characteristics of large surface area to unit volume and small buoyancy. Recently microbubble generators with low energy and high performance are required to wide applications. In the present study, we propose one new effective technique to generate tiny bubbles with less than 200μm diameter utilizing a venturi tube under high void fraction condition. The objective of the present study is to elucidate the mechanism of bubble breakup phenomena in the venturi tube and to clarify the effects of parameters which are necessary to realize an optimum system experimentally. Under low velocity condition, bubbles which were observed with a high speed camera parted gradually in a wide region. On the contrary under high velocity condition, bubbles expanded after passing through the throat and shrank rapidly. Since the speed of sound in gas-liquid system is extremely lower than that of single-phase flow, the bubble breakup phenomenon in the venturi tube is explained as the supersonic flow in a Laval nozzle. With rapid pressure recovery in diverging area, expanding bubbles collapse violently. The tiny bubbles are generated due to the surface instability of shrinking bubbles.

View full abstract

We performed direct numerical simulations (DNS) of turbulent channel flow of surfactant solution to investigate the characteristics of the modulated wall turbulence and the drag-reduction. In the simulations, the effect of the surfactant solution was represented by the modified Bird-Carreau model for the shear viscosity. By comparing with the experimental results, we confirmed that the features of turbulent flow of surfactant solution were reproduced. In fact, one of the results showed the drag reduction rate of 37% and is in the range of low drag reduction. Due to the scale analysis from a viewpoint of instantaneous structures as well as turbulence statistics, we found that the modulation of turbulent flow of surfactant solution from the case of the Newtonian fluid could be generally normalized with the local variable viscosity. Some additional corrections are, however, needed to apply to the prediction of the dissipation rate, which the local viscosity affects directly in turbulent flow.

View full abstract

We have shown the basic equation of the orientational distribution function of prolate spheroidal hematite particles that conduct the rotational Brownian motion in a simple shear flow under an applied magnetic field. The equation has been solved numerically in order to investigate mainly the influence of the spin rotational Brownian motion on the orientational distribution, the negative magneto-rheological effect and the characteristics of magnetization. A prolate spheroidal hematite particle has an important characteristic that it is magnetized in a direction normal to the particle axis. The main results obtained in the present study are summarized as follows. The present results are in good agreement with those of the theory without that motion in that the position and the height of a peak of the orientational distribution and also the whole distribution shape agree significantly well with each other. Hence, it is seen that the spin rotational Brownian motion does not significantly influence the orientational distribution. In contrast, the spin rotational Brownian motion has a quantitative effect on the negative viscosity, although the dependence of the negative viscosity on the magnetic field strength is in qualitatively good agreement with that of the theory without the spin rotational Brownian motion: the effect of the spin rotational Brownian motion quantitatively appears as a significant decrease in the negative viscosity effect. Moreover, the negative magneto-rheological effect can be obtained for a larger aspect ratio or for a more prolate hematite particle. Since the magnetization has a strong relationship with the orientation of the magnetic moment, the effect of the spin rotational Brownian motion appears in this characteristic more significantly.

View full abstract

The flow around a rotating disk in a cylindrical casing is investigated by the numerical method. The radius of the disk is shorter than the inner radius of the casing, as well as the thickness of the disk is smaller than the axial length of the casing. Therefore, the flow field has a radial gap in addition to an axial gap. When there is an axial gap, the flow has been regarded as the cross flow model, and the spiral rolls and spots appear in the Bödewadt layer on the stationary end wall of the casing. Taylor-Couette vortices emerge in the radial gap by the influences of the rotating disk with finite thickness and the side wall of the casing. This results in the generation of the flow with bead-like and sickle-like vortices near the radial gap. We visualize the flow and investigate the appearance and the collapse of bead-like vortices. We also show the transition process of the flows that depend on the Reynolds number. In the transition process with increasing Reynolds number, we can find Taylor vortex flow, flow with bead-like vortices, sickle-like vortices and the spiral rolls. It is shown that the sickle-like vortices is generated only in the case where bead-like vortices is biased to one side of the casing end walls and gives the fluctuation to generate energy and torque.

View full abstract

For the safety operation of a large-scale sodium-cooled fast reactor in Japan (JSFR), structural integrity has to be maintained. Therefore, several thermal-hydraulics phenomena, which may reduce the integrity, have been studied. One of them is the cavitation due to sub-surface vortex which can damage structural surfaces. Such a sub-surface vortex is induced by high-speed suction flow into an outlet pipe in the reactor vessel. However, the onset condition of the cavitation can not be clarified easily because the cavitation shows highly complicated behaviors associated with phase change. Therefore, the authors are developing a CFD code to simulate the cavitation accurately and to investigate the onset condition of the cavitation. In this paper, as the first part of the development, a high-precision simulation scheme for vortical flows is presented. In this scheme, a vortical flow is identified in each cell and a vortical velocity distribution is determined locally to be consistent with the local velocity distribution. Then, the calculations of momentum transport through cell faces are performed in consideration with the vortical velocity distribution. In addition, a pressure distribution near the vortex center is also considered to simulate accurately the mechanical balance between pressure gradient and centrifugal force. As a fundamental verification of the developed scheme, the reproducibility of the burgers vortex is checked. Furthermore, inviscid vortex attenuation in two-dimensional domain is simulated. As a result, it is confirmed that the developed scheme can simulate vortical flows more accurately than conventional schemes.

View full abstract

Bubble jet flows are seen in a lot of industrial fields such as power plant, chemical plant and bio-reactor. PIV is very useful tool in order to investigate the bubble's behaviors and flow structures in liquid phase simultaneously. Because it is possible to measure two- or three-dimensional flow structures and the whole flow fields aren't disturbed by measurement instruments. Recently, Dynamic PIV system consisted of high-speed camera and high-repetition laser has been developed. It is possible to capture time-serial images with high temporal resolution. In this paper, a system combined dynamic PIV and IST (Infrared Shadow Technique) with single high-speed camera, a double-pulse laser for illumination of particles and a diode laser for illumination of bubbles is developed in order to capture particles and bubbles simultaneously. Some techniques about image pre-processing to separate particles and bubbles and velocity calculation by PIV/PTV are introduced. POD (Proper Orthogonal Decomposition) is furthermore applied to time-series liquid phase velocity vectors. It is revealed that the effect of spatio smooth filter is obtained by reconstructed velocity vectors in lower spatio mode by POD and it is changed from large vortex structures to small vortex structures as the spatio mode becomes higher.

View full abstract

A DEM-MPS coupled model for solid-liquid two phase flows is developed. Fluid motion is calculated for locally phase-averaged variables. The present model is applied to a collapse behavior of the assembly of aluminum particles in a rectangular container with and without liquid. The corresponding experiment is performed to validate the present model. The trajectory of each particle is traced by using the PTV (Particle Tracking Velocimetry) measurement. Calculated collapse behaviors agree well with the observed behaviors in the corresponding experiment. The calculated horizontal position of leading particle and particle velocity also agree well with the experimental results.

View full abstract

An Electro-Conjugate Fluid (ECF) is one of functional fluids. A strong jet flow is generated between two electrodes when a high voltage is applied to ECF through the electrodes. By the use of this strong jet flow, a pump with simple structure, no noise and no vibration can be developed. In this view point, authors had developed printed circuit board multi-layered type ECF-pump and tube type ECF-pump in order to realize a new liquid cooling system by ECF. And the performance of the liquid cooling system composed of printed circuit board multi-layered or tube type ECF-pump had been investigated. In this study, an ECF pump using two metallic meshes as electrodes is proposed. The proposed pump is fabricated, and experiments are performed to investigate the basic characteristics of the pump varying mesh size. These experimental results show that the existence of a suitable combination of metallic meshes is made clear and the proposed ECF pump has remarkable properties. In addition, the basic characteristics of this pump are investigated by carrying out experiments varying the width between two metric meshes of this pump. These experimental results show that this pump has remarkable properties when the width between two metric meshes is set at 0.1mm.

View full abstract

In this study, we evaluated the dipteran passive pitching motions caused by the fluid-structure interaction using the dynamically scaled experiments. Since the model wing and the surrounding fluid interact with each other, the dynamic similarity between the model and the actual dipteran flights was measured using not only the Reynolds and Strouhal numbers but also the mass and Cauchy numbers. Using these four numbers, the data of the flapping flights of Tipula obsolete, Tipula paludosa, Eristalis tenax, Calliphora vicina were dynamically scaled and applied to the model wing flights. In all tested insect flights the characteristic pitching motions were well simulated, i.e. the model wing maintained the high angle of attack in the middle of each half stroke and rotated during the stroke reversals. The simulated pitch amplitudes were equivalent to those of observations in the actual insect flights and the generated lift was sufficient to support the insect weight. As a consequence the dipteran pitching motion is fundamentally caused by the fluid-structure interaction.

View full abstract

The conventional Galerkin finite element solution is mesh-dependent and sometimes can not converge on the original Poisson's equation locally when a unstructured linear mesh is used. We try to solve these two problems by introducing a new concept of the virtual nodal domain(Vnd) for a linear tetrahedral element and distributing the source term to a nodal algebraic equation in proportion to the Vnd. The Vnd is evaluated from the conservation law using a second-order flux. We have proved theoretically that the total Vnd belong to the four nodes equals to the volume of the element. Numerical simulation of heat conduction with both Dirichlet and Neumann boundary conditions shows that the accuracy has been improved obviously comparing with the conventional Galerkin FEM for unstructured tetrahedral meshes, especially for bad quality elements.

View full abstract

In this work, we have investigated phonon transport in silicon crystal using molecular dynamics and lattice dynamics methods. The phonon relaxation time is derived from molecular phase space trajectories through two different analysis methods using normal mode projection and spectral energy density. By performing the calculations for wavevectors spanning the entire first Brillouin zone, we find that these two relaxation-time calculation methods give almost the same results despite the fundamental difference in the underlying theories. With the obtained phonon relaxation time and group velocity calculated by lattice dynamics method, we have quantified the contribution from each phonon mode to the overall thermal conductivity. In addition, by calculating the cumulative thermal conductivity, we have quantified contributions to thermal conductivity from phonons with different mean free paths to gain insight into the scale effect of heat conduction in nanoscale.

View full abstract

A computational model was developed to predict the pyrolysis process of municipal solid waste in externally heated rotary kiln. The model was calculated assuming solid waste and gas are one-dimensional flow. The heat transfer coefficient among solid waste, gas and inner cylinder of kiln, which is variously affected by kiln structure and operation condition, was acquired by a scale model examination. A flexible computational grid was adopted in the model, which could change its shape according to the height of solid waste layer in each time step. The accuracy of this model was verified by comparison with the pilot scale experiments and the actual plant data. The predicted transient temperature profile is in good agreement with experimental data. Compared to the actual plant data, temperature prediction error is less than 10% at inner cylinder wall, pyrolysis gas, and solid waste.

View full abstract

For better understanding of soot formation and oxidation processes in a bio-diesel spray flame, morphology, microstructure and size of soot particles directly sampled in a spray flame fuelled with soy-methyl ester were investigated using high-resolution transmission electron microscope (HRTEM). The soot samples were taken at different axial locations in the spray flame, 40, 50 and 70 mm from the injector nozzle, which correspond to soot formation, peak, and oxidation zones, respectively. The bio-diesel spray flame was achieved in a constant volume combustion chamber under a diesel-like high pressure and temperature condition (6.7 MPa, 1000 K). Density, diameter of primary particles and radius of gyration of soot aggregates directly sampled onto the TEM grids in the flame reached a peak at 50 mm from the injector nozzle and was lower or smaller at 40 and 70mm.

View full abstract

In order to evaluate evaporation heat fluxes from coolant water in a spent fuel pit of a nuclear power plant to ventilation air during a shutdown of water purification and cooling systems, empirical correlations were derived. To derive correlations, the evaporation heat transfer databases at Shinsyu University, which were obtained using test sections A and B with heat transfer lengths of 940mm and 300mm, were used. The temperatures of the hot water and air were 35-65°C and about 20°C, respectively, and air velocity was up to 2.08m/s. In this study, a correlation including length scale was derived using the database under the outlet relative humidity less than 100% (X_out < 1.0) in the test section B and the analogy between heat and mass transfers. The heat flux data with 100% relative humidity at the outlet (X_out = 1.0) in the test section A were corrected using the heat flux data with X_out < 1.0 in the test section B in order to obtain evaporation heat fluxes under 100% relative humidity (X = 1.0) conditions, which might be applied to a spent fuel pit with the length scale of about 10m. Then, another correlation without the length scale was introduced from the heat fluxes corrected for conditions of X = 1.0. The heat fluxes for the length scale of 10m calculated using the two correlations agreed each other.

View full abstract