Alkaline water electrolysis with magnetic field is a method for increasing the efficiency of hydrogen production through the electrolysis of alkaline water solutions. In the author’s research group, experiments and simulations were conducted to elucidate the multiphase flow behavior and electrolytic efficiency in alkaline water electrolysis with magnetic field. It was demonstrated that the electrolytic efficiency can be improved by changing the bubble motion, even with electrodes that are operable with permanent magnets (without the need for magnet power) and easily scalable, and with a weak magnetic field applied vertically.

This article outlines the fundamentals of electric conducting fluid analysis and its numerical model for an electrically conductive droplet when alternating magnetic field is applied. The level-set method was applied for interfacial capture and the electromagnetic potential was introduced for electromagnetic field analysis. The pressure and electromagnetic potential were computed by the Red-Black SOR method using GPU (Graphics processing unit). It is exhibited that the solid, liquid and gas phases were able to be computed simultaneously and that an electromagnetic levitation phenomenon was successfully simulated.

In the continuous casting process, argon gas is injected into the tundish to homogenize the molten steel and remove the undesired inclusion. Furthermore, the molten steel flow is controlled by applying a magnetic field to the mold. The inclusion of argon gas as the molten steel flows into the mold can lead to product defects. Therefore, it is essential to predict the motion of the bubble in liquid metals under influence of the magnetic field, and further investigations are required for the liquid-metal two-phase flows. However, measurement techniques which can be applicable to opaque fluids such as liquid meat are limited. The authors have developed a high-speed ultrasonic tomography for measuring bubbles motion and investigated the behavior of continuously released bubbles motion in gallium alloy under influence of a horizontal magnetic field. The ultrasonic tomography measurements were carried out with 500 frames per second. It was shown that as the magnetic field strength increases, the variation of the bubble passing position in the measurement cross-section becomes smaller. Furthermore, it was confirmed that the distribution changed slightly depending on the direction of the applied magnetic field and that the bubbles tended to distribute in the direction perpendicular to the magnetic field. This trend changes depending on the difference in the gas flow rate. It is thought that the greater the influence of the wake, the more pronounced the anisotropy of the distribution.

Contactless sample manipulation is attracting great attention in a wide range of fields, including analytical chemistry, biological science, and food industry. A better understanding on the droplet dynamics in acoustic levitation is vital to realize the lab-in-a-drop technology. Here, this paper reviews the recent research progress on the contactless droplet manipulation driven by acoustic levitation, focusing on droplet levitation, transport, coalescence, mixing, and evaporation dynamics are presented and discussed.

The pressure measurement using a fiber optic prove hydrophone and simultaneous two-directional observation with a high-speed video camera have been conducted to investigate the collapse of a laser-induced bubble near a rigid wall and the corresponding shockwave generation. Pressure measurements at multiple locations near the rigid wall and observations of frontal and overhead views relate to the locations of pitting damage due to bubble collapse observed by Philipp and Lauterborn (J. Fluid Mech., 1998). In γ=s/R_max≅1.5 where s is the distance between the bubble center and the wall surface and R_max is the equivalent bubble radius at the maximum growth, the second bubble collapse occurs on the rigid wall: the overhead view reveals that the bubble collapses asymmetrically with a toroidal (ring) shape and bubble blobs remain at the bubble center and on both sides of the ring in the direction of laser injection. It is shown that the area of the bubble blobs on the ring correlates with the high pressure on the wall. This suggests that the collapse of these blobs leads to the high pressures and asymmetric pitting damages on the wall.

A microbubble generated by the laser beam’s photothermal conversion induces convective flow by the Marangoni effect at the gas-liquid interface. This flow is useful for the transport and accumulation of nanoparticles in liquid. This study investigates the regime of the convective flow in nanoparticle suspension around the generated microbubble. The empirically observed flow regime was initially similar to that of theoretically predicted convection in pure water, which has a circulating flow driving particles 100 μm away from the microbubble. However, after 10 ms of the laser heating, the size of the circulating flow decreased to the same as the bubble diameter. Additionally, a numerical simulation also shows that the decreasing of the circulating flow is observed when the driving force at the interface is limited to the shear stress of > 140 N/m².

In decommissioning Fukushima Daiichi nuclear power station, the issue is confinement of radioactive aerosols in the primary containment vessel. Although a HEPA (High Efficiency Particulate Air) filter is used to collect the aerosol particles, pretreatment equipment such as a scrubber may be applied to reduce the load of HEPA filters. In the scrubber, the aerosol particles are removed by moving from gas to liquid through gas-liquid interface. Since the CE (collection efficiency) depends on gas-liquid interfacial area, fine bubbles are necessary to obtain high collection efficiency. JAEA (Japan Atomic Energy Agency) developed a new particle removal technique by using bubble breakup phenomenon in a Venturi tube. To confirm usefulness of the technique, we performed the CE measurements and observed gas-liquid two-phase flow in the Venturi tube. In comparison with a straight pipe type, the Venturi type can have removed particles more 1,000 than it. The CE is almost the same as a HEPA filter. In addition, the Venturi type has the enough CE as the pretreatment equipment for various materials of particles such as Kanto loam, SUS and oil. Besides, we clarified that the CE of the Venturi type depended on the gas and liquid flow rates. The CE increases with the increase of the liquid flow rate but decreases with the increase of the gas flow rate. This is because the CE is affected by the bubble breakup phenomenon in the Venturi tube. In the experiment, we confirmed that cavitation number which is the parameter of the bubble breakup was related to the CE of the Venturi type.

In this study, antimicrobial effect of mixed ultra-fine bubble liquids on two types of bacteria; spore-forming bacteria (Bacillus subtilis var. natto) and the environmental bacteria was investigated. We used ultrapure water (UPW), mixed ultra-fine bubble water (UFB), mixed ultra-fine ozone-rich bubble water (UFORB), and water containing ozone (OZ). Viable bacteria count (VBC) of UFB agreed with that of UPW within the experimental errors (p < 0.05). Low VBC (High effect) of UFORB and OW exhibited on B. subtilis var. natto. Moreover, complete sterilization of UFORB was obtained on the environmental bacteria. For discussing the experimental results, we observed number density against particle diameter, effect of dilution and elapsed time after stopped the ultra-fine bubble generator. Difference between UFB and UFORB exhibited in case of distribution of particle diameter (peak diameter and averaged diameter). The bactericidal effect on B. subtilis var. natto persisted after 360 min (6 h), but disappeared after 1440 min (24 h). On the other hand, the bactericidal effect against the environmental bacteria was high even after 1440 min (24 h).

Gas-liquid two-phase flow is a very complicated flow phenomenon that involves the interaction between gas and liquid phases. Recently, an accurate simulation technique of two-phase flow has been developed. However, the validation of the simulated results is insufficient due to less experimental data. It is especially difficult to measure interfacial behavior with significant spatio-temporal fluctuation. To understand such phenomena, measurement methods with high spatial and temporal resolutions are required. Neutron imaging is a powerful tool for two-phase flow visualization. It can be applied to flow measurement in an opaque channel and has applicability to the heated condition. In this study, two-phase flow in the narrow rectangular channel was visualized by high-speed neutron imaging. The dynamics of the interfacial structure of air-water flow were investigated from the neutron transmission images, and the liquid film behavior was observed qualitatively.

In the safety study on a sodium-cooled fast reactor, it is assumed that the high-temperature fuel debris are formed because of a core disruptive accident. During the cooling of the fuel debris, gas-liquid two-phase flow can be generated in the debris due to the coolant boiling. To improve the prediction accuracy of the flow characteristics, detailed measurement of the gas-liquid two-phase flow in the debris bed is required. In this study, gas-liquid two-phase flow in a quasi-two-dimensional sphere-packed bed, which simulates the debris bed, is visualized by using X-ray imaging. The experimental results show that the local void fraction increases near the splitting section in the packed bed and decreases near the coalescence section.

To clarify the thermal aspect of cavitating flow, cavitation experiments were conducted using a heated nozzle with different mainstream temperatures and heating conditions to evaluate two thermal effects that affect the cavity. The decrease in cavity length was observed due to the thermodynamic self-suppression effect by increasing the mainstream temperature, and the cavity length increased by heating from the wall. By comparing the variation of cavity length in the non-heated and heated conditions, the condition balancing the two thermal effect was found. Additionally, the response of each thermal effect to the mainstream temperature was also found. It was also shown that the required wall superheat to cancel out thermodynamic self-suppression effect reaches a maximum value when the mainstream temperature is around 70 °C. The experimental results were compared and discussed with the mathematical model values for the two thermal effects.

The final goal of this study is to develop a “data-driven cavitation model” based on a machine learning model that is trained by the measurement data, rather than a conventional mathematical model, aiming at improvement of the model by the data assimilation aspect of using measured data. In this paper, as a preliminary study, a framework for a data-driven cavitation model is developed by employing CFD data calculated with a mathematical homogeneous fluid model as the training dataset. The target is cavitating flow around a Clark-Y11.7% hydrofoil at an angle of attack of 0 degree. The neural network for data-driven cavitation model is U-Net, which is a kind of convolutional neural network. The training dataset consists of the fields of the velocity, pressure, and liquid volumetric fraction f_L of the current timestep and those of the next timestep. In the test mode, trained U-Net is generally able to predict the f_L field of the next step except at the trailing edge vortex where the cavity rapidly grows. The data-driven cavitation model implemented in the flow simulation code quantitatively reproduces the same flow field as that of the mathematical model. However, computation becomes unstable where cavity grows rapidly and pressure waves are generated due to cavity volume fluctuations, suggesting that a data-driven cavitation model needs further improvements to robustly capture such highly unsteady phenomena.

In the cleaning and coating processes of product manufacturing, some cases require filling liquid into closed-end holes. We have developed a liquid infiltration method, i.e., gas removal, into closed-end holes by irradiating acoustic waves. The method uses gas oscillation inside the holes; however, the final removal process of break-up gas columns is still unclear. This study observed the multiple gas column oscillation inside a closed-end hole during acoustic wave irradiation and its removal process. We also visualized the liquid flow between columns using the PIV technique. As a result, we found that the gas removal occurrence depends on the size, distance, and the number of gas columns. The removal process shows that the gas columns were approached and coalesced. They were then ejected through capillary wave propagation, necking, break-up, and re-coalescence. The oscillation of the gas column near the hole exit induced the flow between columns and approaching the other gas column. The PIV analysis shows that the long-distance gas column cases induced smaller flow velocities and could not achieve gas removal due to liquid inertia.

The simulation of gas-liquid two-phase flow through a porous media by using multiple GPUs have performed. The ideal parallel efficiency of the weak scaling is achieved by applying weakly compressible method, which can avoid to solve Poisson's equation. We can calculate a wide range of parameters by introducing weakly compressible method. The conservative Allen-Cahn equation is solved to capture the gas-liquid interface by using the finite volume method to conserve the total mass of each phase. The simulations of gas-liquid two-phase flow through two-dimensional porous medium are performed by using our solver by using multi GPUs. Ideal parallelization efficiency is achieved by overlapping computation and communication time.

This study deals with a numerical simulation of the growth of bubble nuclei and the corresponding bubble cloud formation in a pressure field given by high intensity focused ultrasound (HIFU) backscattered from a laser-induced bubble interface. In the present paper, using a multiscale numerical method (Tamura et al., Japanese J. Multiphase Flow, 2022) in which the ghost fluid method is coupled with bubble dynamics, the growth of the bubble cloud which consists of multiple bubble layers is simulated continuously as observed in the experiments. It is shown that the distance between the first cavitation inception point and a laser-induced bubble interface is about 0.27λwhere λ is the wavelength of HIFU; the distance between bubble layers in the cloud is also about 0.35λ these distances in the simulation are in good agreement with the experiments. The influence of the configuration of bubble nuclei on the final form of a cone-shaped bubble cloud is investigated. It is also shown that the shape of each bubble layer, which depends on the initial configuration of bubble nuclei, affects the final formation of bubble clouds. This is because that the negative pressure due to the backscattering of HIFU is dependent on the shape of each bubble layer. The results also show that the averaged final shape of bubble clouds quantitively matches the results as observed in the experiment by Horiba et al. (J. Acoust. Soc. Am., 2020).

The void fractions of large and small bubbles, α_G,L and α_G,S, in a bubble column were measured by means of the DGD (Dynamic Gas Disengagement) method. The column diameter was 200 mm and the initial liquid heights H₀ were 800, 1600, 2400, and 3000 mm. We dealt with heterogeneous bubbly flows with the superficial gas velocity J_G from 0.06 to 0.30 m/s. Three kinds of liquids were used as the liquid phase to understand effects of the liquid viscosity on the void fractions. The following conclusions were obtained: (1) an experimental database of the void fractions of large and small bubbles was obtained at various initial liquid heights and liquid viscosities, (2) the void fractions of large and small bubbles increase with increasing J_G at constant H₀, (3) the void fraction of large bubbles for J_G ≤ 0.10 m/s decreases with increasing H₀ for H₀ < 2400 mm and is almost constant for H₀ ≥ 2400 mm while α_G,L is independent of H₀ for J_G ≥ 0.20 m/s, (4) the liquid viscosity does not affects α_G,L, (5) the void fraction of small bubbles slightly increases at J_G = 0.06 m/s, whereas it is almost constant at J_G = 0.10 m/s, and decreases for J_G ≥ 0.20 m/s as H₀ increases, and (6) the void fraction of small bubbles decreases with increasing the liquid viscosity and dependence of α_G,S on H₀ is not so different.

Passive production of synthesis gas from liquid methanol using a capillary action in a packed bed of porous particles supporting a catalyst is investigated. Heating of an upper side of the packed tube while the bottom is immersed in liquid methanol causes upward fluid flow due to capillary action enhanced by evaporation and produces the synthesis gas in the high temperature region. For an increased diameter of the bed, the partial invasion of wetted region into the core of the dried region may cause a significant reduction in reaction yield. The radial distribution of temperature and its effect on the reaction are identified from the experiment. Un-uniformity caused in the radial direction is discussed based on the physical modeling of the capillary liquid flow toward the heated wall.