Two-phase flow loop cooling system is required for increasing heat transfer rate, heat transport distance, and cooling heat flux in thermal control of space structures. The understandings on thermo-fluid dynamics of gas-liquid two-phase flows with phase change under microgravity is necessary for the system design. To clarify the phenomena and verify the operation of two-phase flow loops under microgravity condition, two-phase flow loop experiments were conducted as a JAXA project named TPF experiment onboard the International Space Station (ISS). The loop was a pump-driven loop and the pressure was maintained by an accumulator. Perfluorohexane was used as the working fluid because of low latent heat of vaporization and moderate boiling point under the restriction of power input and for the safe operation. The loop had two kinds of heating section, such as a copper heating tube and a transparent glass heating tube, and a condenser. Flow observation sections made of transparent polycarbonate resin were installed just downstream of each heating tube in order to observe and measure the interface structure of boiling two-phase flow in detail. In this report, the design concept and system characteristics of the experimental loop and the results of the gas-liquid two-phase flow measurements in the flow observation section are presented.

This paper discusses the transient model of loop heat pipes in detail. A new model was constructed using an implicit method for fluid calculations and an explicit method for heat transfer calculations between pipe walls and radiators. In addition, the effectiveness of this model was verified using a loop heat pipe used for demonstration purposes on International Space Station. Good agreement was shown with the results of thermal vacuum tests. On the other hand, a comparison with the on-orbit test results revealed differences in vibration phenomena. It was concluded that this was because the heat input in orbit, such as solar heat input, was not accurately estimated.

Fluid behavior in microgravity is different from in ground gravity since surface tension, viscous force, and wetting are dominant in this condition. In propellant tank for artificial satellite and future on-orbit spacecraft, sloshing due to disturbance and settling behavior by change of acceleration have to be understood for design of propellant supply system and attitude control system. These fluid behaviors in microgravity condition are affected by a dynamic wetting significantly, therefore it is important to understand dynamic wetting which dominates fluid behavior. We observed transient fluid behaviors in cylindrical containers upon entering microgravity conditions created by drop tower facility, and effect of viscosity and diameter of container on fluid behaviors were investigated. CFD (Computational Fluid Dynamics) analyses were conducted and these results were compared with experimental results. Additionally, it was confirmed that numerical model considering dependence of contact line velocity in dynamic contact angle by empirical and theoretical methods provided more reasonable results. In addition, the sloshing behaviors inside the cylindrical containers were investigated under applied horizontal and axial accelerations using both microgravity experiments and CFD. As a result, it was found that under conditions of low excitation acceleration and high viscosity coefficient of the liquid, where the fluid movement is minimal, the experimental and numerical analysis results exhibit good agreement.

Long-term storage of cryogenic propellants such as liquid hydrogen and liquid oxygen is required for manned long-distance exploration of deep space. In propellant tank, evaporation from the gas-liquid interface is accelerated by sunlight and heat input from surrounding equipment, and the boil-off gas (BOG) generated from the liquid surface, it is caused pressure raise in the tank and could lead to tank destruction. Therefore, the development of Thermodynamic Vent System (TVS) to reduce BOG generation is urgently needed. An effective way to reduce BOG is to agitate the bulk liquid by mixing cold jet issuing from the bottom of the tank. Here, to estimate the jet-tip height, one-dimensional dynamic model is described. And the height of the jet-tip is investigated under various gravitational conditions, and the experimental results compared to the results obtained from the dynamic model.

Rotational separations in binary droplet collision are simulated using the improved lattice Boltzmann method for incompressible two-phase flows. Firstly, a couple of interpolation methods for the fluid density and viscosity in gas-liquid interfaces are examined, and the linear interpolation for the density and the harmonic interpolation for the viscosity are applied to the simulations of rotational separation. The calculated results are classified into three types of collision, i.e., coalescence, stretching separation and rotational separation in the We'-B diagram, where We' is the effective Weber number defined by using the calculated relative velocity immediately before the collision and B is the impact parameter. It is found that the rotational separation is observed in the region around B=0.4 for 20<We'<30. Secondly, the coalesced droplet length after collision is investigated, and time evolutions of the length are calculated for the rotational separations. The results indicate that the rotational separation may have different mechanism from that of the stretching separation. Thirdly, colored tracer particles are calculated to investigate the flow characteristics inside the droplets, and it is found that each droplet after rotational separation includes some of the other droplet. This outcome is more significant in the case of B=0.46 than B=0.42. Finally, the mixing intensity is defined to investigate the mixing of the droplets. The results showed that the droplets in the case of B=0.42 are more mixed than the case of B=0.46, because the contact surface area in the former case is larger than that in the latter case.

As one of the severe accident countermeasures, a method of injecting gas into pool water in a suppression chamber to capture and decontaminate aerosol particles contained in the gas is proposed. For severe accident countermeasures, a physical model for single bubbles representing hydrodynamic phenomena is included. The aim of this study is to elucidate the effect of aerosol particles on the shape and upward behavior of single large diameter bubbles. For this purpose, several aerosol particles with different solubility in tap water were used as parameters to investigate the effect of bubble diameter and Eötvös number on large single bubbles. As a result, the aspect ratio of single bubbles with aerosol particles is smaller than that of droplets without aerosol particles. Furthermore, the aspect ratio was smaller when containing soluble aerosol particles than insoluble aerosol particles. The terminal velocity and drag coefficient were less affected by aerosol particles in the region where bubble diameter is large. Comparison of the numerical and experimental results for aspect ratio showed a similar trend in the increase of aspect ratio with the increase of Eötvös number, but for the quantitative values, the experimental results were much lower than the numerical results.

Because of investigating effect of mixed ultra-fine ozone-rich bubble (UFORB) liquids on the degradation of organic compounds, we observed molar concentration of aqueous solutions of methylene blue (organic compound model) against elapsed time by using UFORB or ozone-water (OW) suspended in ultrapure water (UPW) or tap-water (TAP). Excellently lower concentration of methylene blue by using UFORB exhibited. Compared with that of OW, effect of both high decomposing ability and short processing time was excellently obtained. For discussing the experimental results, additional two experiments were conducted. Dissolved ozone-concentration of UFORB suspended in UPW or TAP was measured. For case of UFORB, highly ozone-concentration in ultrapure water exhibited and was independence of elapsed time. However, that in tap-water decreased against elapsed time within the experimental errors. Relationship between number density and particle diameter of ultra-fine bubble was measured. Peak diameter of UFORB-TAP was smaller than that of UFORB-UPW in relatively initial times. Moreover, differences between stability of ultra-fine bubble and ozone-concentration suspended in UPW or TAP was discussed.

Heat transfer of the stratified/wavy water flow in a horizontal tube is investigated. Under uniform heating, considerably low heat transfer in the vapor phase region relative to the liquid phase region produces a large circumferential temperature distribution, which makes useless to apply the averaged heat transfer coefficient over the perimeter for the estimation of the heat transfer area of heat exchangers. From the distributions of the actual surface heat flux evaluated experimentally, it is confirmed that the heat circumferentially conducted through the tube over the vapor phase region does not affect the liquid phase region due to the high heat transfer in the proximity of the phase boundary. The heat transfer coefficient in the liquid phase is regarded as a representative of the heat transfer of the stratified/wavy flow and predicted by accounting for the contributions of convection and evaporation and compared with measurements.

Heat transfer of the stratified/wavy water flow in a horizontal tube is investigated. Considerably low heat transfer in the vapor phase region relative to the liquid phase region produces a large temperature distribution along the perimeter, which makes difficult for deriving the appropriate expression for the heat transfer coefficient. In the 1st report, it was shown that, based on the experiment, the heat transfer coefficient in the liquid phase could be a representative of the entire tube. A model which involved both of convection and evaporation was proposed, where a flat surface of the liquid layer was assumed. In the 2nd report, the spread of the liquid layer due to wetting with the tube surface and its effects on the heat transfer coefficient were accounted for. The prediction agreed with the measured results within approximately 30% error, which is more accurate compared to the prediction for a flat surface.