Direct simulations are performed for a deformable bubble rising steady through a quiescent viscous fluid for different Reynolds and Weber numbers. The axi-symmetric Navier-Stokes equations for the liquid phase are solved on a boundary-fitted mesh, which allows an accurate prediction of the bubble shape and the drag force acting on the bubble. The drag coefficients are in excellent agreement with previous experimental data, while the deviation of the Moore's theory from these numerical and experimental results is found to increase with decrease of the Reynolds number and with increase of the bubble aspect ratio. It is revealed that neither the fore-aft asymmetry nor the presence of a standing eddy at the rear of the bubble is of primal importance to the drag force.
Contribution of a toroidal vortex behind a bubble to its steady drag was experimentally investigated. Bubble motion in eight different liquids were observed using high-speed photography. Photochromic dye was used to visualize the rear vortex structure. We clarified that the critical aspect ratio of bubble obtained previously in numerical simulations for the onset of the existence of the standing eddy and three-dimensional wake behind an oblate spheroid bubble corresponds well to our experimental results when the front aspect ratio is used. Non-dimensional steady drag on a bubble was found to have a linear relation with the front aspect ratio. It is also confirmed that the existence of standing eddy hardly changes the drag.
An effective numerical method for the prediction of the convective mass transfer from a dispersed gaseous phase is proposed. To enhance the prediction capability for the advection of the concentration as well as the volume fraction for the bubbly flow, the mixed Youngs-centered method and the Lagrangian-explicit scheme are applied together with the volume-of-fluid method. The dissolution of gas at the gas-liquid interface is modeled based on the Henry's law, and is implemented by an overwriting method for the mass transfer in the liquid. The validation of the present method is performed for the two-dimensional advection problem and the one-dimensional diffusion problem. Preliminary results on the mass transfer from freely rising bubbles in an initially quiescent liquid are also demonstrated.
An experimental investigation about the interaction among successive large bubbles in the vertical upward flow was conducted. In the experiment, single and two/three consecutive rising bubbles were artificially made, and the velocities of large bubbles flow were measured by using the CECM (Constant Electric Current Method) and a high speed camera. In the case of two consecutive bubbles, the maximum velocity was influenced by the volume of the leading bubble. In the case of three consecutive bubbles, the former two bubbles, i.e. “Top bubble” and “Middle bubble”, showed the same characteristics with two successive bubbles, but later two bubbles, i.e. “Middle bubble” and “Bottom bubble”, showed the different tendency with two successive bubbles. The effect of the difference especially was observed in the deformation of the bottom bubbles.
The shrinking and growth of microbubbles under pressure change are observed with a CCD camera. The influence of gas diffusion on the stability of microbubbles coated by phospholipid is investigated. The microbubbles are made with acoustic liposomes encapsulating phosphate buffer solution and perfluoropropane gas. It is shown that when the ambient liquid pressure increases, the observed microbubbles shrink accompanied with the cyclic surface deforming and smoothing process. The bubble surface smoothing shows that the excess phospholipid layers are removed from the surface, which results in the instantaneous bubble shrinkage. It is also shown that the smaller the initial radius is, the more the growth of microbubbles is reduced. The experiments are compared with the simulations in which the dynamic surface tension and the variation of gas permeation resistance of molecular layers are taken into account. The instantaneous bubble shrinkage in the experiment is simulated successfully with the bubble model in which the instantaneous increase of surface tension due to the reduction of the excess phospholipid layer material is taken into account.
Helically coiled tubes are utilized for heat exchanger in many cryogenic industries. The helically coiled tube shows the different flow mechanism with vertical tubes, i.e. secondary flow and phase separation. In order to understand the boiling heat transfer characteristics in a helically coiled tube, experiment was conducted by using a forced convective boiling system with liquid nitrogen. Test section had 4 mm in inner tube diameter, 100 mm in coil diameter and 3150 mm in heated length. The experimental conditions were as follows; mass flux was 200-710 kg/m2s and system pressure was 0.3 MPa. As experimental results, the pressure drop, boiling heat transfer coefficient and critical heat flux had been evaluated including the post-dryout region.
The gas-liquid flow distributions in the multi-pass upward channels that simulated the compact evaporators for the automobile air-conditioning system were investigated experimentally. Special attention was directed to influences of the inlet flow pattern at the header entrance, pressure condition at the branch outlets, and pressure-loss characteristics of branches on the gas-liquid distributions. Experiments were conducted in an isothermal air-water flow system. It was found that the pressure condition at the branch outlets exerted a great influence on the gas-liquid distributions to branches when flat smooth tubes were adopted as branches, but it had a minor influence when the branches consisted of multi-port tubes. The inlet flow pattern at the header entrance had a significant influence on the gas-liquid distribution, and liquid tended to be distributed to all the branches when the inlet flow pattern is a mist flow.
For the optimum design of two-phase flow nozzle of CO2 two-phase flow ejector, it is important to investigate fluid mechanics of high-speed gas-liquid two-phase flow. The purpose of this study is to investigate CO2 two-phase nozzle flow in view of both experimental and analytical aspects. In the experiment, CO2 in supercritical pressure condition was blow down into atmosphere through a divergent-convergent nozzle, where the mainstream profile of temperature and the jet force at nozzle exit were measured. In the analysis, one-dimensional model which assumes steady, adiabatic, frictionless, and equilibrium was proposed. The flow in the nozzle of convergent part was treated as single-phase flow, whereas the flow in the nozzle of divergent part was treated as separated two-phase flow with saturation. Although this analysis is quite simple, the analytical results can follow the experimental results well within the cases in this study.
As the two-phase flow regime evolves from bubbly to slug flows along the flow direction in a vertical large diameter pipe, the void fraction increases first in the bubbly flow region, then decreases in the transitional region, and finally increases again in the slug flow region. The N-shaped profile of void fraction was numerically analyzed by using the one-dimensional two-fluid model. Its interfacial drag model was modified by including constitutive equations of C0 and Vgj developed by Shen et al.  for evolving two-phase flow from bubbly to slug flows in a large diameter pipe. The numerical calculation predicted well the N-shaped profile of the measured axial void fraction distribution from the present experiments. The predicted average gas and liquid velocities indicated that the average gas velocity increased significantly and the average liquid velocity decreased slightly when the void profile is N-shaped one. The phenomenon can be attributed to the formation and development of large bubbles of pipe size.
Microbubble formation at a T-shaped microfluidic junction with gas pressure change was experimentally investigated. The gas pressure was actively controlled by using an audio speaker. With this method of bubble generation, we are able to control the size and generation interval of bubble independently. The microchannel has a rectangular cross section of 0.1mm×0.1mm, and channel length of 2.5mm from the junction. As results, we found that rise in gas pressure leads to the elongation of a gas column, the length of which is proportional to both the amplitude and cycle of gas pressure oscillation. The decrease in gas pressure then shortens the gas column. During this shortening period, the necking motion of the column and the following micro bubble generation were successfully achieved at the T-shaped junction.
Experiments were made to clarify the mean void fraction, the mean holdup, the velocity of liquid slug and the flow patterns of air-water two-phase flow in horizontal flat rectangular microchannels with the dimensions of the width and depths of 1.0 x 0.1 mm and 1.0 x 0.2 mm, respectively. The flow patterns of bubble flow, slug flow and annular flow were observed. These data in the microchannel showed relatively similar tendencies to that in the minichannels with the width of 1~10mm and the depth of 1mm which we had previously reported. In a channel of 1.0 x 0.1 mm, however, the mean holdup and the holdup which correspond to the base film thickness in annular flow gave larger values because the effects of liquid viscosity and surface tension on the holdup and void fraction became dominant. The remarkable flow characteristics of rivulet flow and the flow with partial dry out mainly on the upper channel wall were observed in slug flow and annular flow regions in the channel of 0.1 mm depth.
An optical measurement system was used to investigate the effect of microchannel length on adiabatic gas-liquid two-phase flow characteristics. Experiments were conducted with 1,676 mm long, circular microchannel with the diameter of 100 μm. Two-phase flow patterns, void fraction and velocities of gas and liquid slugs were measured at any location from gas-liquid mixer to channel exit. The experimental values of the mean void fraction and the mean velocity of liquid slug agreed well with their values given by the equation of the homogeneous flow model from gas-liquid mixer up to channel exit when the flow rate of the liquid was constant and the mass velocity of the gas was lower. The transition of the flow patterns from slug flow to ring film flow was observed when the flow rate of the liquid was constant and the mass velocity of the gas became large.
Gas-liquid two-phase flows in porous structures are simulated by the lattice Boltzmann method for two-phase fluids with large density differences, in which the wetting boundary condition on solid walls is incorporated. The behavior of the liquid phase penetrating the hydrophobic and hydrophilic microchannels is investigated. The curvature radius and the average height of the liquid phase in equilibrium are calculated, and found to be in good agreement with the theoretical predictions. In addition, the behavior of the liquid phase in porous structures with various porosities is simulated for hydrophilic and hydrophobic cases. From these results, it is found that the liquid phase in the porous structure is transported from the hydrophobic to hydrophilic regions, and that the liquid tends to penetrate in locally narrow spaces owing to the effect of large pressure gradient.
In this article, natural convection of a temperature-sensitive magnetic fluid in a porous media is studied numerically by using lattice Boltzmann method. Results show that the heat transfer decreases when the ball numbers increase. When the magnetic field is increased, the heat transfer is enhanced; however the average wall Nusselt number increases at small ball numbers but decreases at large ball numbers due to the induced flow being more likely confined near the bottom walls with a high number of obstacles.
Temperature sensitive magnetic fluid (TSMF) has a kinetic characteristic of temperature dependence on magnetization in an applied magnetic field. In the present study, a binary magnetic fluid, which is a mixture of the TSMF and an organic secondary fluid, is utilized in a heat transport device to enhance its circulation. In order to verify the performance of the TSMF mixture in the heat transport device, the driving characteristic of this heat transport device with 45° inclination of the heating section is investigated both experimentally and numerically. Results showed that the driving force is enhanced greatly due to the magnetic field intensity increase and the effect of gas bubbles.
Liquid gallium jet was injected from glass capillary into stationary water layer and the breakup behavior was observed in detail, as a simulation study of the metal-particle production process. Atmosphere, oxidative or de-oxidative, was controlled by applying DC voltage between jet and surrounding water. In de-oxidative atmosphere, the jet broke up almost similar manner to the water jet into air. However, in oxidative atmosphere, the jet showed unusual breakup manner and the oxide deposited inside the nozzle in some cases. Based on the results, the mechanisms of several thwart problems, which occurred in the particle production process of melt jet breakup, were discussed. The measures against the problems were also suggested in this paper.
The defocus method is known as a method which can measure three dimensional coordinates of particles by one camera, and it is especially suitable for the 3D-PTV measurement of micro-flows. Although the color defocus method has solved some problems of the conventional defocus method, light absorption by color filters caused new difficult problems of insufficient light intensity. This paper proposes two new techniques which can take pictures more clearly than the existing color defocus method. By employing light-color filters, the serious problem of light absorption can be resolved. As a result, the combined use of those filters and HSI color analysis allows wider selection of filter colors. The introduction of laser illumination brings wider applications of the color defocus method, which enforces measurement precision of the method. Some measurement examples are shown here.
Authors have developed a new particle tracking velocimetry using path-line images without respite of exposure. A new detection method of coordinate of end-points path-lines is proposed in this paper. It is achieved by adding hue to the information of grayscale image captured by two cameras. Validation of the method is examined using artificial images. Further an application result of the method to measurement of 3D fluid motion induced by ascending spheres is presented.
A simple algorithm using 4-sensor probe was derived to measure bubble velocity in a gas-liquid two-phase flow. Measurement error of bubble velocity and shape was estimated numerically and experimentally, assuming spherical and ellipsoidal bubble. The result shows that the bubble velocity can be measured within 10% error at 0 < jl <1.0 [m/s] by using this method.
Numerical approaches of the solid-liquid flows were not established so far. This is because the modeling of free surface and solid phase was difficult, furthermore the calculation cost might become excessive. In the present study, a new method is developed to simulate the solid-liquid flows involving the free surface. In this method, the solid-liquid flows were computed by combining the Discrete Element Method (DEM) and the Moving Particle Semi-implicit (MPS) method. This is called the DEM-MPS method. In the present study, the validation of the DEM-MPS method was performed in a solid-liquid flow involving free surface in a rotating tank. The angle of repose and solid distribution were compared between the simulations and experiments. The simulation results were in good agreement with those obtained by the experiment.
Using the Immersed Boundary Method with Physical Virtual Model proposed by Lima e Silva et al., characteristics of unidirectional flow through circular cylinders is investigated. Calculation result on a pressure drop through circular cylinders is compared with the empirical formula by Ergun. Transition of flow pattern through circular cylinders with increase in the Reynolds number is presented. In addition, spatial and time variation of fluid force acting on each cylinder is presented.
It is known that a cavitating jet causes a ring-shaped erosion distribution on an impingement wall. A purpose of the present study is to make clear the mechanism of high impact and erosion due to cavitation cloud. We made some observations of cavitation cloud and erosion using an oblique wall with asymmetrical state of cavitation cloud. We tried to correlate the erosion distribution with behaviors of cavitation cloud using an image analysis about an oblique impingement type of cavitating water jet. As a result, a characteristic erosion distribution and a collapsing behavior of cavitation cloud appear on the test plate. The phenomena of pressure wave propagation with cloud collapses are shown quantitatively using an image analysis in relation to the erosion distribution. It is shown that a chain reaction of cloud collapses and propagation of pressure wave play an important role in the case of oblique impingement of cavitating water jet.
We performed three experiments using the spray of steam-water mixture. The first one is the experiment of removing dry-etch residue from a wafer. We found that the spray can remove the dry-etch residue in via holes with the help of a small amount of chemicals. This result is in sharp contrast with the result with the use of air-water mixed spray, and suggests that the removal of dry-etch residue is essentially due to physical actions of the spray, rather than the actions of chemicals. The second and third experiments are resist lift-off and metal erosion, respectively. We found that the performances of both depend on the water flow rate in the same manner; a certain amount of water is helpful, but excessive amount of water is of no effect. Based on these experimental results, we propose that in cleaning or eroding surfaces using the steam-water mixed spray, the condensation effect of steam plays an important role and the performances can be controlled by tuning the amount of water.