Study on premixing phase of steam explosion at JAERI

Melt jet breakup and fragmentation has been studied in ALPHA program at JAERI. In the first two experiments of the MJB series, a jet of molten lead–bismuth eutectic alloy was released into a deep pool of saturated water. The steam generation rate was measured and correlated with the jet behavior observed by a high-speed camera. The jet breakup length and debris size distribution were also evaluated. In parallel with the experimental study, JASMINE code has been developed for the simulation of a steam explosion process. The models of melt jet breakup and the particle breakup in the code were assessed by analyzing FARO-L14 and ALPHA MJB experiments.     

Numerical analysis of fragmentation mechanisms in vapor explosions

Fragmentation of molten metal is the key process in vapor explosions. However, this process is so rapid that the mechanisms have not yet been clarified in experimental studies. In addition, numerical simulation is difficult because we have to analyze water, steam and molten metal simultaneously with boiling and fragmentation. The authors have been developing a new numerical method, the moving particle semi-implicit (MPS) method, based on moving particles and their interactions. Grids are not necessary. Incompressible flows with fragmentation on free surfaces have been calculated successfully using the MPS method. In the present study, numerical simulation of the fragmentation processes using the MPS method is carried out to investigate the mechanisms. A numerical model to calculate boiling from water to steam is developed. In this model, new particles are generated on water–steam interfaces. A two-step pressure calculation algorithm is also developed. Pressure fields are separately calculated in both heavy and light fluids to maintain numerical stability with the water and steam system. The new model and algorithm are added to the MPS code. Water jet impingement on a molten tin pool is calculated using the MPS code as a simulation of collapse of a vapor film around a melt drop. Penetration of the water jet, which is assumed in Kim–Corradini’s model, is not observed. If the jet fluid density is hypothetically larger, the penetration appears. Next, impingement of two water jets is calculated. A filament of the molten metal is observed between the two water jets as assumed in Ciccarelli–Frost’s model. If the water density is hypothetically larger, the filament does not appear. The critical value of the density ratio of the jet fluid over the pool fluid is ρ_jet/ρ_pool=0.7 in this study. The density ratios of tin–water and UO₂–water are in the region of filament generation, Ciccarelli–Frost’s model. The effect of boiling is also investigated. Growth of the filament is not accelerated when the normal boiling is considered. This is because normal boiling requires more time than that of the jet impingement, although the filament growth is governed by an instant of the jet impingement. Next, rapid boiling based on spontaneous nucleation is considered. The filament growth is markedly accelerated. This result is consistent with the experimental fact that the spontaneous nucleation temperature is a necessary condition of vapor explosions.     

Simulation of melt jet breakup experiments by JASMINE with an empirical correlation for melt particle size distribution

We modified JASMINE code, a fuel–coolant interaction simulation code developed at Japan Atomic Energy Agency (JAEA), to extend the applicability for ex-vessel melt coolability assessment. The modification included addition of a melt particle size distribution model based on an empirical correlation and a simple non-local radiation heat transfer model, improvement in the treatment of melt particle generation, and re-agglomeration of settled particles. The modified code was tested by simulating melt jet breakup experiments, namely selected cases of ALPHA/GPM series with alumina–zirconia mixture and steel melt by JAEA, and FARO experiments with urania–zirconia mixture by Joint Research Center Ispra. Simulation results showed that the code reproduces the experimental results well for the cases with a deep subcooled water pool where the melt breaks up completely. On the other hand, significant underestimation of heat removal from the melt and overestimation of agglomeration of settled melt was encountered for conditions with a shallow or saturation temperature water pool. The melt agglomeration behavior in the simulation was sensitive to model parameters on the agglomeration criterion and heat transfer depending on conditions.     

Development of the evaluation methodology for the material relocation behavior in the core disruptive accident of sodium-cooled fast reactors

The in-vessel retention (IVR) of core disruptive accident (CDA) is of prime importance in enhancing safety characteristics of sodium-cooled fast reactors (SFRs). In the CDA of SFRs, molten core material relocates to the lower plenum of reactor vessel and may impose significant thermal load on the structures, resulting in the melt-through of the reactor vessel. In order to enable the assessment of this relocation process and prove that IVR of core material is the most probable consequence of the CDA in SFRs, a research program to develop the evaluation methodology for the material relocation behavior in the CDA of SFRs has been conducted. This program consists of three developmental studies, namely the development of the analysis method of molten material discharge from the core region, the development of evaluation methodology of molten material penetration into sodium pool, and the development of the simulation tool of debris bed behavior. The analysis method of molten material discharge was developed based on the computer code SIMMER-III since this code is designed to simulate the multi-phase, multi-component fluid dynamics with phase changes involved in the discharge process. Several experiments simulating the molten material discharge through duct using simulant materials were utilized as the basis of validation study of the physical models in this code. It was shown that SIMMER-III with improved physical models could simulate the molten material discharge behavior, including the momentum exchange with duct wall and thermal interaction with coolant. In order to develop an evaluation methodology of molten material penetration into sodium pool, a series of experiments simulating jet penetration behavior into sodium pool in SFR thermal condition were performed. These experiments revealed that the molten jet was fragmented in significantly shorter penetration length than the prediction by existing correlation for light water reactor conditions, due to the direct contact and thermal interaction of molten materials with coolant. The fragmented core materials form a sediment debris bed in the lower plenum. It is necessary to remove decay heat safely from this debris bed to achieve IVR. A simulation code to analyze the behavior of debris bed with decay heat was developed based on SIMMER-III code by implementing physical models, which simulate the interaction among solid particles in the bed. The code was validated by several experiments on the fluidization of particle bed by two-phase flow. These evaluation methodologies will serve as a basis for advanced safety assessment technology of SFRs in the future.     

A simple novel and fast computational model for the LIVE-L4

The LIVE-L4 test was conducted to investigate the transient and steady state behavior of the molten pool and the crust influenced by different heat generation rates. The main purpose of this work is to develop a simple novel model of the LIVE code to calculate the entire process of the LIVE-L4 test after the melt of KNO₃–NaNO₃ poured into the test vessel. The LIVE code is a transient code and can be used as a fast computational program to calculate the LIVE tests. Natural convection heat transfer in the melt pool, crust behavior, heat conduction in the vessel wall, and radiative heat transfer were all considered in the model of the LIVE code.In the LIVE code, Asfia–Dhir correlations were used to calculate average and local heat transfer coefficients in the melt pool. With the assumption of no considering the composition change of local melt at melt/crust interface, many important parameters, including the melt pool temperature, heat flux distribution along the vessel wall, the thickness of the crust in steady state, and crust growth rate during the test, were calculated and compared with the LIVE-L4 experimental data.The melt pool Nu calculated by the LIVE code is larger than experimental data due to the use of Asfia–Dhir correlation in the LIVE code, which caused the average heat flux through the vessel wall larger than experiment data except the heating phase of 5 kW. It is attributed that the temperature difference between the melt pool temperature and the interface temperature at melt/crust measured in the test is larger than that calculated by the LIVE code due to the constant interface temperature at melt/crust of 284 °C used in the LIVE code. Crust growth rate calculated by the LIVE code was consistent well with the experiment data. Calculation results indicated that the LIVE code could generally predict the main parameters of the melt and crust well during the LIVE-L4 test.     

Study effects of the kinetics of local composition diffusion on the melt pool coolability by developing a DBL-LIVE model

The LIVE-L4 test was conducted to investigate the transient and steady state behavior of the molten pool and the crust influenced by different heat generation rate. In previous work, a simple novel model of the LIVE code was developed to simulate the entire process of the LIVE-L4 test after the melt of KNO₃NaNO₃ poured into the test vessel. The LIVE code is a transient code and can be used as a fast computational program to simulate the LIVE tests. Calculation results indicated that the LIVE code could generally predict the main parameters of the melt and crust well during the LIVE-L4 test.However, the LIVE code could not predict some processes accurately, such as the early phase of the test after the melt poured into the test vessel, and heat flux distribution at small polar angle, due to no considering the composition change of local melt during the crust formation, which affected the properties of the local melt adjacent to the crust, especially the liquidus temperature of the local melt at the interface of melt and crust. Therefore, considering the effects of the crust growth rate and the diffusion of concentrated NaNO₃ on the composition change of the local melt, the diffusion boundary layer (DBL) model was developed and combined with the LIVE code model to consider the composition change of the local melt during the crust formation, with the assumption of considering a quasi-stable plane front solidification of the melt. The DBL-LIVE code was used to calculate the characteristics of the melt and the crust during the heating phase 18 kW. The calculation results indicated that the DBL-LIVE code could predict the main parameters of the melt and crust much more accurately. It could also predict the transient and steady characteristics of the melt composition. Moreover, the thickness of the diffusion boundary layer has a marked influence on the composition distribution. Therefore, the sensitivity analysis of the DBL thickness was also conducted in this work.     

Experimental and analytical studies of melt jet-coolant interactions: a synthesis

Instability and fragmentation of a core melt jet in water have been actively studied during the past 10 years. Several models, and a few computer codes, have been developed. However, there are, still, large uncertainties, both, in interpreting experimental results and in predicting reactor-scale processes. Steam explosion and debris coolability, as reactor safety issues, are related to the jet fragmentation process. A better understanding of the physics of jet instability and fragmentation is crucial for assessments of fuel-coolant interactions (FCIs). This paper presents research, conducted at the Division of Nuclear Power Safety, Royal Institute of Technology (RIT/NPS), Stockholm, concerning molten jet-coolant interactions, as a precursor for premixing. First, observations were obtained from scoping experiments with simulant fluids. Second, the linear perturbation method was extended and applied to analyze the interfacial-instability characteristics. Third, two innovative approaches to computational fluid dynamics (CFD) modeling of jet fragmentation were developed and employed for analysis. The focus of the studies was placed on (a) identifying potential factors, which may affect the jet instability, (b) determining the scaling laws, and (c) predicting the jet behavior for severe accident conditions. In particular, the effects of melt physical properties, and the thermal hydraulics of the mixing zone, on jet fragmentation were investigated. Finally, with the insights gained from a synthesis of the experimental results and analysis results, a new phenomenological concept, named ‘macrointeractions concept of jet fragmentation’ is proposed.     

Molten aluminum alloy fuel fragmentation experiments

Experiments were conducted in which streams of molten aluminum alloys were injected into a 1.2 m deep pool of water. The parameters varied were (i) injectant material (8001 aluminum alloy and 12.3 wt% U-87.7 wt% Al), (ii) melt superheat (0 to 50 K), (iii) water temperature (313, 343 and 373 K) and (iv) size and geometry of the pour stream (5, 10 and 20 mm diameter circular and 57 mm annular). The pour stream fragmentation was dominated by surface tension with large particles ( 30 mm) being formed from varicose wave breakup of the 10-mm circular pours and from the annular flow off a 57 mm diameter tube. The fragments produced by the 5 mm circular jet were smaller ( 10 mm), and the 20 mm jet which underwent sinuous wave breakup produced 100 mm fragments. The fragments froze in 313 K water to form large solid particles with high voidage which would be readily coolable. However, in water ≥ 343 K the melt fragments did not freeze during their transit through 1.2 m of water and agglomerated into a melt pool at the bottom of the vessel.     

Fragmentation behavior during molten material and coolant interactions

A safe design for a fast breeder reactor (FBR) requires post-accident heat removal (PAHR) for any potential core disruptive accident (CDA). It is important to ensure that the molten core material solidifies in the sodium coolant in the reactor vessel even if all of the core material has melted. In the present experiment, molten material was injected into water to experimentally obtain the information on the molten material jet entering the coolant and its fragmentation. Visual information was obtained with a high-speed video camera, showing that fragmentation behavior on the side of the jet was different from that on the jet front, and that the injection nozzle diameter significantly influenced the jet breakup length, while the molten jet temperature and the coolant temperature did not influence the jet breakup length. Comparison of the diameters of fragments of the solidified molten material thus obtained with fragmentation theory shows that the median fragment diameter is between the critical Weber number theory and the most-unstable wavelength of the instability theory of surface waves at a gas liquid interface.The quench behavior of the molten jet in coolant was calculated for FBR conditions by using the model that reflects actual fragmentation behavior. It was clarified that the mass of molten material in the coolant pool is related to the fragment diameter under FBR conditions.     

A numerical and experimental study of water ingression phenomena in melt pool coolability

During a postulated severe accident, the core can melt and the melt can fail the reactor vessel. Subsequently, the molten corium can be relocated in the containment cavity forming a melt pool. The melt pool can be flooded with water at the top for quenching it. However, the question that arises is to what extent the water can ingress in the corium melt pool to cool and quench it. To reveal that, a numerical study has been carried out using the computer code MELCOOL. The code considers the heat transfer behaviour in axial and radial directions from the molten pool to the overlaying water, crust generation and growth, thermal stresses built-in the crust, disintegration of crust into debris, natural convection heat transfer in debris and water ingression into the debris bed. To validate the computer code, experiments were conducted in a facility named as core melt coolability (COMECO). The facility consists of a test section (200 mm × 200 mm square cross-section) and with a height of 300 mm. About 14 L of melt comprising of 30% CaO + 70% B₂O₃ (by wt.) was poured into the test section. The melt was heated by four heaters from outside the test section to simulate the decay heat of corium. The melt was water flooded from the top, and the depth of water pool was kept constant at around 700 mm throughout the experiment. The transient temperature behaviour in the melt pool at different axial and radial locations was measured with 24 K-type thermocouples and the steam flow rate was measured using a vortex flow meter. The melt temperature measurements indicated that water could ingress only up to a certain depth into the melt pool. The MELCOOL predictions were compared with the test data for the temperature distribution inside the molten pool. The code was found to simulate the quenching behaviour and depth of water ingression quite well.     

大型先进压水堆熔融物堆内滞留初步研究

参考国外熔融物堆内滞留(IVR)稳态包络工况计算编写相关程序,并与ERI、DOE及INEEL的结果进行比较,对程序进行验证。通过对大型先进压水堆熔池参数和不确定性分析可知,如果使用ULPU-2000台架Ⅳ的流道设计,压水堆发生超CHF事故的可能性小于7%,但压力容器壁厚最大熔化量超过15 cm的可能性很大,如果没有其他缓解措施,建议将大型先进压水堆压力容器厚度增加至20 cm以上。热流分配是影响熔池行为的主要因素,建议采取措施调整熔融池热流分配,以缓解氧化物层和金属层交界面处的传热危机。     

Thermal Fragmentation of a Molten Metal Jet Dropped into a Sodium Pool at Interface Temperatures below Its Freezing Point

In order to verify the thermal fragmentation of a molten jet dropped into a sodium pool at instantaneous contact interface temperatures below its freezing point, a basic experiment was carried out using molten copper and sodium. Twenty grams of copper was melted in a crucible with an electrical heater and was dropped through a 6 mm nozzle into a sodium pool of 553 K, in the form of a jet column. Thermal fragmentation originating inside the molten copper jet with a solid crust was clearly observed in all runs. It is verified that a small quantity of sodium, which is locally entrapped inside the molten jet due to the organized motion between the molten jet and sodium, is vaporized by the sensible heat and the latent heat of molten copper, and the high internal pressure causes the molten jet with a solid crust to fragment. It is also found that the fragmentation caused in the molten copper-sodium interaction was severer than that in the molten uranium alloy jet-sodium interaction, which was reported by Gabor et al, under the same superheating condition and lower ambient Weber number condition of molten copper.     

Simulation of melt jet breakup and debris bed formation in water pools with IKEJET/IKEMIX

The objective of the development of the code system KESS is simulating the processes of core melting, relocation of core material to the lower head of the reactor pressure vessel (RPV) and its further heatup, modelling of fission product release and coolability of the core material. In the scope of the code development, IKEJET and IKEMIX were designed as key models for the breakup of a molten jet falling into a water pool, cooling of fragments and the formation of particulate debris beds. Calculations were performed with these codes, simulating FARO corium quenching experiments at saturated (L-28) and subcooled (L-31) conditions, as well as PREMIX experiments, e.g. PM-16. With the assumption of a reduced interfacial friction between water and steam as compared to usually applied laws, the melt breakup, energy release from the melt and pressurisation of the vessel observed in the experiments are reproduced with a reasonable accuracy. The model is further applied to reactor conditions, calculating the relocation of a mass of corium of 30 t into the lower plenum, its fragmentation and the formation of a particle bed.     

Experimental investigation of jet fragment diameter in BWR complicated structures using image processing techniques

ABSTRACT

To estimate the status of the Fukushima Daiichi nuclear power plant’s reactor pressure vessel, it is important to understand the breakup and fragmentation behavior of a molten jet of core fuel in the lower plenum of a boiling water reactor (BWR). To clarify the effects of complicated structures on jet breakup and fragmentation, we conducted experiments to visualize jet falling behavior, simulating conditions of a severe BWR accident using a multi-channel experimental apparatus. In this study, we developed a new image-processing method that could recognize fragments in a liquid-liquid flow and estimate fragment diameter of a molten jet in the BWR lower plenum under several inlet conditions. We clarified that complicated structures, such as control rod guide tubes, have little effect on fragment diameter, and that inlet velocity is the dominant factor affecting fragment diameter. This indicates that shear stress would occur at the crest of interfacial waves at the side of the jet, leading to its fragmentation. Finally, the fragment diameters measured in this study show good agreement with the correlation based on shear stress, and a new correlation to predict the jet fragment diameter was developed by fitting the constant correlation value from the experimental result using a simulant fluid.     

Study on fragmentation behavior of liquid lead alloy droplet in water

Fragmentation behavior of molten lead alloys droplet in water was investigated experimentally by releasing liquid LBE (45w%Pb-55w%Bi) and lead droplets into a pool of subcooled water. The fragmentation occurred when the temperature of the interface between a molten droplet and water was higher than the spontaneous nucleation temperature of water and lower than the minimum film boiling temperature. With increasing the droplet temperatures, the peak pressure in fragmentation of LBE droplet increased from 5 to 8 kPa, and for lead, the value remained around 2 kPa. With increasing the water subcooling, the peak pressure in fragmentation remained constant at 5 kPa for LBE droplet and at 2 kPa for lead droplet. The lead alloy fragmentation process in water was numerically simulated by embedding a semi-empirical fragmentation model for droplet fragmentation rate into the computer code of two-phase flow: JASMINE code. The corresponding results, such as pressure history and fragmentation peak pressure, agreed well with the experimental results.     

Small-scale Experiment on Subcooled Water Jet Injection into Molten Alloy by Using Fluid Temperature-Phase Coupled Measurement and Visualization

The plunging water jet behavior into a pool of a molten lead-bismuth alloy is experimentally investigated. The mixing and interactions of fluids were detected by measuring the fluid temperature as well as the fluid phase distinction by the newly developed bifunctional probes.

In general knowledge of fuel-coolant interactions, the film boiling of water is caused immediately after the first contact of high-temperature melt and water, but the vapor film is locally collapsed by some reasons and the direct contact is extensively propagating in some cases which may produce the explosive vapor generation. In the melt-injection mode previously investigated by numerous researchers, the triggering of explosive interactions is considered as a local rewetting caused by instabilities of the vapor film as the melt temperature decrease. In the coolant-injection mode discussed by the present study, on the other hand, the water temperature poured into bulk melt continues to rise for penetration, in general, that should be effective to stabilize the film boiling. The present experiments showed, however, that the explosive boiling occurred in a condition that both water and melt initial temperatures were high enough for maintaining stable film boiling on the melt-water interface that is clearly different manner of the melt injection mode. Such unstable phenomena are observed when the instantaneous interfacial contact temperature exceeds the homogeneous nucleation temperature of water and the amount of saturated water is accumulated in a melt pool.     

Natural convection heat transfer test for in-vessel retention at prototypic Rayleigh numbers – Results of COPRA experiments

Large-scale COPRA experiments were performed to investigate the natural convection heat transfer in melt pools for the in-vessel retention during severe accidents in Chinese large-scale advanced PWRs. Both water and binary mixture of 20 mol% NaNO₃ – 80 mol% KNO₃ were used as the melt simulant material in performed tests. Due to the full scale geometry of the COPRA test section, the Rayleigh numbers of the melt pool could reach up to the prototypic magnitude of 10¹⁶. Natural convection heat transfer tests at prototypic Rayleigh numbers have been performed to study the influence of the heat generation rate and melt simulant material on the melt pool temperature, heat flux distribution and heat transfer capability. The comparisons of the melt pool temperature and heat flux distribution from water experiments and molten binary salt experiments showed that the crust formation along the inner surface of the vessel wall could impact the heat transfer characteristics of the melt pool. And the heat flux distribution from COPRA water tests and molten salt tests were in good agreement with those from Jahn-Reineke water experiments and RASPLAV molten salt experiments, respectively. The heat transfer capability of the melt pool Nu_dn from COPRA molten salt tests were larger than those from water tests, but both were lower than those from ACOPO and BALI predictions within the same range of Rayleigh numbers (10¹⁵ – 10¹⁷).     

A numerical simulation of water jet injection behavior in fuel–coolant interaction

Water injection mode of molten fuel and coolant interaction is a key issue during the steam generator tube rupture accident in liquid metal reactors. The focus of the present study is placed on the numerical simulation of the water jet behavior falling into a pool of a denser fluid in order to get qualitative and quantitative understanding of initial premixing phase of water injection mode. A multi-phase code with the volume of fluid (VOF) method is developed. The simulation results are compared with experimental data to examine the capability of the current approach. Effects of density ratio and Froude number on cavity penetration velocity are quantitatively analyzed. The simulation results show surface waves and breakup behavior occur both at the top of the cavity during cavity collapse and at the cavity boundary. The simulation results are compared with the existing theories. At the top of the cavity, the water jet wavelength is close to the value estimated based on the Rayleigh–Taylor instability. At the cavity boundary, melt wave length is close to the value estimated based on the Kelvin–Helmholtz instability.     

Experimental study on breakup and fragmentation behavior of molten material jet in complicated structure of BWR lower plenum

To estimate the state of reactor pressure vessel of Fukushima Daiichi nuclear power plant, it is important to clarify the breakup and fragmentation of molten material jet in the lower plenum of boiling water reactor (BWR) by a numerical simulation. To clarify the effects of complicated structures on the jet behavior experimentally and validate the simulation code, we conduct the visualized experiments simulating the severe accident in the BWR lower plenum. In this study, jet breakup, fragmentation and surrounding velocity profiles of the jet were observed by the backlight method and the particle image velocimetry (PIV) method. From experimental results using the backlight method, it was clarified that jet tip velocity depends on the conditions whether complicated structures exist or not and also clarified that the structures prevent the core of the jet from expanding. From measurements by the PIV method, the surrounding velocity profiles of the jet in the complicated structures were relatively larger than the condition without structure. Finally, fragment diameters measured in the present study well agree with the theory suggested by Kataoka and Ishii by changing the coefficient term. Thus, it was suggested that the fragmentation mechanism was mainly controlled by shearing stress.     

Effect of melt water interaction configuration on the process of steam explosion

Steam explosion experiments are performed at various modes of melt water interaction configuration using prototypic corium melt. The tests are performed to simulate both melt water interaction in a partially flooded cavity and melt water interaction in a cavity with submerged reactor. The tests are performed using zirconia and corium melts. The behavior of melt jet fragmentation during the flight in the air and fragmentation and mixing of melt jet in water is investigated by a high-speed video visualization and by comparison of debris size distribution and morphology of debris. Strength of steam explosion is estimated by measuring dynamic pressure and dynamic force.