Molten Fuel-coolant Interaction Research Articles

The “suevite” conundrum, Part 1: The Ries suevite and Sudbury Onaping Formation compared

AbstractThe term “suevite” has been applied to various impact melt‐bearing breccias found in different stratigraphic settings within terrestrial impact craters. Suevite was coined initially for impact glass‐bearing breccias from the Ries impact structure, Germany, which is the type locality. Various working hypotheses have been proposed to account for the formation of the Ries suevite deposits over the past several decades, with the most recent being molten‐fuel‐coolant interaction (MFCI) between an impact melt pool and water. This mechanism is also the working hypothesis for the origin of the bulk of the Onaping Formation at the Sudbury impact structure, Canada. In this study, the key characteristics of the Ries suevite, the Onaping Formation and MFCI deposits from phreatomagmatic volcanic eruptions are compared. The conclusion is that there are clear and significant lithological, stratigraphic, and petrographic observational differences between the Onaping Formation and the Ries suevite. The Onaping Formation, however, shares many key similarities with MFCI deposits, including the presence of layering, their well‐sorted and fine‐grained nature, and the predominance of vitric particles with similar shapes and lacking included mineral and lithic clasts. These differences argue against the viability of MFCI as a working hypothesis for genesis of the Ries suevite and for a required alternative mechanism for its formation.

Water-magma interaction and plume processes in the 2008 Okmok eruption, Alaska

Eruptions of similar explosivity can have divergent effects on the surroundings due to differences in the behavior of the tephra in the eruption column and atmosphere. Okmok volcano, located on Umnak Island in the eastern Aleutian Islands, erupted explosively between 12 July and 19 August 2008. The basaltic andesitic eruption ejected ∼0.24 km3 dense rock equivalent (DRE) of tephra, primarily directed to the northeast of the vent area. The first 4 h of the eruption produced dominantly coarse-grained tephra, but the following 5 wk of the eruption deposited almost exclusively ash, much of it very fine and deposited as ash pellets and ashy rain and mist. Meteorological storms combined with abundant plume water to efficiently scrub ash from the eruption column, with a rapid decrease in deposit thickness with distance from the vent. Grain-size analysis shows that the modes (although not their relative proportions) are very constant throughout the deposit, implying that the fragmentation mechanisms did not vary much. Grain-shape features consistent with molten fuel-coolant interaction are common. Surface and groundwater drainage into the vents provided the water for phreatomagmatic fragmentation. The available water (water that could reach the vent area during the eruption) was ∼2.8 × 1010 kg, and the erupted magma totaled ∼7 × 1011 kg, which yield an overall water:magma mass ratio of ∼0.04, but much of the water was not interactive. Although magma flux dropped from 1 × 107 kg/s during the initial 4 h to 1.8 × 105 kg/s for the remainder of the eruption, most of the erupted material was ejected during the lower-mass-flux period due to its much greater length, and this tephra was dominantly deposited within 10 km downwind of the vent. This highlights the importance of ash scrubbing in the evaluation of hazards from explosive eruptions.

Investigation of fragmentation phenomena and debris bed formation during core meltdown accident in SFR using simulated experiments

The event of a severe core melt down accident, resulting in the relocation of the active core is analyzed as a part of the nuclear reactor safety research in order to ensure safe removal of decay heat. Molten Fuel Coolant Interaction (MFCI) and debris bed configuration on the core catcher plate assumes importance in assessing the post accident heat removal capability. The key factors affecting the coolability of the debris bed are the bed porosity, morphology of the fragmented particles, degree of spreading/heaping of the debris on the core catcher and the fraction of lump formed. A well defined debris bed is helpful in fixing a prototypical source term for the PAHR studies. Towards this, a series of experiments on fragmentation kinetics and subsequent debris bed formation is conducted with molten Wood's metal (an alloy of Bi 50%, Pb 25%, Sn 12.5% and Cd 12.5% with melting point of 346K) in water simulant system. The experiments are carried out using 2kg, 5kg and 20kg melt inventories. The particle size distribution obtained for the fragmented debris is fit using an Upper Limit Log Normal (ULLN) distribution. The dependence of particle size distribution on initial melt temperature and interaction height is quantified by correlating them to the key parameters i.e. shape factor and location factor in the ULLN expression. Morphology of the debris particles is investigated to understand the fragmentation mechanisms involved. Three major mechanisms of fragmentation are identified namely melt entrainment mechanism, boundary layer stripping and hydrodynamic breakup due to capillary forces. Finally, an approach to quantify the stratification and spreading behaviour of debris on the collector tray is presented by analysing the particle size and mass percentage in different radial zones of the collector tray.

Prototypic corium oxidation and hydrogen release during the Fuel–Coolant Interaction

The article is focused on the hydrogen release by chemical reaction of Fuel and Coolant (FCR) during the molten Fuel–Coolant Interaction (FCI) and its impact on the steam explosion (SE) efficiency. Samples of debris coming from several FCI tests have been characterized by solid-state analytical techniques. Two kinds of tests have been analyzed: (i) Experiments using simulant materials (MISTEE, PREMIX, ECO FCI experiments) and tests using prototypic corium i.e., in the U–Zr–O system (KROTOS experiments). The nonradioactive melts produce during FCI from negligible amounts of hydrogen up to 3.6g per kg of melt. It was found that UO2–ZrO2 melt solidifies during FCI into face-centered (Fm3m) crystal cell forming a U1−xZrxO2+y solid solution. The oxygen over-stoichiometry of the solid solution is the evidence of FCR and release of hydrogen. A method combining chemical analysis by Energy Dispersive X-ray Spectrometry (SEM/EDS) and X-ray Powder Diffraction (XRD) allowed the determination of the oxygen over-stoichiometry y. Therefore, it has been indirectly estimated how much hydrogen was formed during the chemical reaction between hot corium and water. These data are provided for KROTOS KS2, KS4 and KS5 tests performed in the frame of the OECD/NEA Serena phase-2 project. It has been shown that up to i.e., 4.7g of H2 per kg of delivered corium melt can be formed by the reaction stoichiometric or sub-oxidized UO2–ZrO2 melts with water. The oxygen over-stoichiometry of the U1−xZrxO2+y debris (initially x=0.48 for KS2; x=0.35 for KS4; x=0.38 and y=−0.37 for KS5) varied between 0.107⩽y⩽0.167. The hydrogen formation during FCI is primarily linked to the melt composition and secondary to the size of the fragmented melt droplets.

Trigger - and heat-transfer times measured during experimental molten-fuel-interactions

A modified setup featuring high speed high resolution data and video recording was developed to obtain detailed information on trigger and heat transfer times during explosive molten fuel-coolant-interaction (MFCI). MFCI occurs predominantly in configurations where water is entrapped by hot melt. The setup was modified to allow direct observation of the trigger and explosion onset. In addition the influences of experimental control and data acquisition can now be more clearly distinguished from the pure phenomena. More precise experimental studies will facilitate the description of MFCI thermodynamics.

Analysis of material effect in molten fuel–coolant interaction, comparison of thermodynamic calculations and experimental observations

Molten Fuel–Coolant Interaction (FCI) is considered an important phenomenon in the frame of nuclear reactor severe accident research and development. Recently, the so-called “material effect” in FCI was observed in several experimental research programs. It was found that chemically different melts yield to different explosion energetics or trigger-ability during FCI tests. The presented work is focused on the thermodynamic modeling of high temperature chemical reactions during the melt–coolant interaction and consequent comparison with results obtained by material analyses of debris coming from KROTOS (CEA, France), MISTEE (KTH, Sweden) and PREMIX (FZK, Germany) experimental research programs. Calculations at thermodynamic equilibrium were obtained using the Gemini 2 (Thermodata) Gibbs energy minimizer. Information about debris composition, morphology and structure were collected using solid-state characterization techniques such as scanning electron microscopy, X-ray powder diffraction and thermogravimetry. Chemical reactions of melt and water vapor were theoretically predicted and experimentally confirmed, and therefore, features like melt solidification path, hydrogen production and melt oxidation are discussed.

Rhyolitic phreatomagmatism explored: Tepexitl tuff ring (Eastern Mexican Volcanic Belt)

Rhyolitic Tepexitl tuff ring in the Serdán-Oriental Basin of the eastern Trans-Mexican Volcanic Belt is a young, well-preserved circular crater — a simple morphology that belies a more complex eruptive history. Field observations, ash studies (granulometry, componentry, and morphology) and volatile data (LOI and XRF) allow the broad division of the Tepexitl deposits into two sequences: a lower sequence characterized by fine-grained deposits with abundant bomb and lapilli sags, and an upper sequence of coarse-grained deposits that show little to no deformation. The early eruptions at Tepexitl are interpreted to have been dominated by discrete, highly efficient, phreatomagmatic blasts, which caused a progressive deepening of the eruptive center (lower sequence), followed by a transition to dominantly magmatic behavior in the upper sequence. Dome growth occurred at the end of the eruption, but subsequent retrogressive explosions triggered by external water destroyed all trace of the original dome morphology. Molten fuel-coolant interaction (MFCI) is commonly thought to cause the repetitive water–magma interaction in phreatomagmatic eruptions, but a process by which magma and viscous felsic magmas can mingle prior to explosive interaction has not been described. A viable mechanism for rhyolitic MFCI, based upon field work at Tepexitl and laboratory experiments, requires that fluidized sediments intrude marginal fractures in the rhyolite magma, creating enough interfacial surface area to initiate phreatomagmatic explosions from within the interior of a rising plug or dome.

Explosive lava–water interactions I: architecture and emplacement chronology of volcanic rootless cone groups in the 1783–1784 Laki lava flow, Iceland

To determine the relationships between rootless cone emplacement mechanisms, morphology, and spatial distribution, we mapped the Hnuta and Hrossatungur groups of the 1783–1784 Laki lava flow in Iceland. We based our facies maps on Differential Global Positioning System (DGPS) measurements, photogeological interpretations, and supporting field observations. The study area covers 2.77 km2 and includes 2216 explosion sites. To establish the timing of rootless cone formation we incorporated tephrochronological constraints from eighty-eight stratigraphic sections and determined that the Hnuta and Hrossatungur groups are composite structures formed by the emplacement of six geographically and chronologically discrete domains. Rootless eruptions initiated in domain 1 on the first day of the Laki eruption (June 8, 1783) and lasted 1–2 days. The second episode of rootless activity began in domain 2 on June 11 and lasted 1–3 days. The four domains of the Hrossatungur group dominantly formed after June 14 and exhibit a complex emplacement sequence that reflects interactions between the Laki lava, contemporaneously emplaced rootless cones, and an existing topographic ridge. In the study area, we identify three distinct rootless cone archetypes (i.e., recurring morphological forms) that are related to tube-, channel-, and broad sheet lobe-fed eruptions. We assert that emplacement of lava above compressible substrates (e.g., unconsolidated sediments) may trigger rootless eruptions by causing subsidence-induced flexure and failure of the basal crust, thereby allowing molten lava (fuel) to come into direct contact with groundwater (coolant) and initiating analogs to explosive molten fuel–coolant interactions (MFCIs).

TRIGGERING AND ENERGETICS OF A SINGLE DROP VAPOR EXPLOSION: THE ROLE OF ENTRAPPED NON-CONDENSABLE GASES

The present work pertains to a research program to study Molten Fuel-Coolant Interactions (MFCI), which may occur in a nuclear power plant during a hypothetical severe accident. Dynamics of the hot ...

Fragmentation of a Single Molten Metal Droplet Penetrating Sodium Pool I Copper Droplet and the Relationship with Copper Jet

The progression of hypothetical core disruptive accidents in metallic fuel fast breeder reactors is strongly affected by the fragmentation of molten metallic fuels due to the molten fuel-coolant interaction (FCI). As a basic study of FCI, the present paper focuses on the fragmentation of a single molten copper droplet with mass from 1 to 5 g, whichpenetrated a sodium pool at instantaneous contact interface temperatures (Ti) from 995 to 1,342°C. Intensive fragmentation of a single molten copper droplet was clearly observed even if Ti values are below the melTsing point (1,083°C) of copper besides the higher Ti range. The intensive fragmentation shows that the mass median diameters (Dm) of copper droplets with a fivefold difference in mass or the same mass have little difference, i.e., they are nearly the same. Under the lower Ti condition, the Dm data of droplet fragments of both the same and different masses scatter widely. It is found that the present Dm/D0 data of mass median diameter normalized by the diameter before touching sodium (D0) give a distribution with larger values than those of molten copper jets with large mass from 20 to 300 g under the lower Ti condition, which were previously reported by the authors, because of the limited amount of heat of droplets. The present Dm=D0 data under the higher Ti condition are found to show an effecTive fragmentation compared with those of molten copper jets with a large mass of 4 kg.

Fragmentation of a Single Molten Metal Droplet Penetrating Sodium Pool I Copper Droplet and the Relationship with Copper Jet

The progression of hypothetical core disruptive accidents in metallic fuel fast breeder reactors is strongly affected by the fragmentation of molten metallic fuels due to the molten fuel-coolant interaction (FCI). As a basic study of FCI, the present paper focuses on the fragmentation of a single molten copper droplet with mass from 1 to 5 g, whichpenetrated a sodium pool at instantaneous contact interface temperatures (Ti) from 995 to 1,342°C. Intensive fragmentation of a single molten copper droplet was clearly observed even if Ti values are below the melTsing point (1,083°C) of copper besides the higher Ti range. The intensive fragmentation shows that the mass median diameters (Dm) of copper droplets with a fivefold difference in mass or the same mass have little difference, i.e., they are nearly the same. Under the lower Ti condition, the Dm data of droplet fragments of both the same and different masses scatter widely. It is found that the present Dm/D0 data of mass median diameter normalized by the diameter before touching sodium (D0) give a distribution with larger values than those of molten copper jets with large mass from 20 to 300 g under the lower Ti condition, which were previously reported by the authors, because of the limited amount of heat of droplets. The present Dm=D0 data under the higher Ti condition are found to show an effecTive fragmentation compared with those of molten copper jets with a large mass of 4 kg.

Phreatomagmatic explosions of rhyolitic magma: Experimental and field evidence

Experimental studies on explosive molten fuel‐coolant interaction (MFCI) using basaltic melt compositions and water as the coolant have provided insight into the physical processes of basaltic and andesitic phreatomagmatic volcanism. Abundant field evidence indicates that rhyolitic and dacitic phreatomagmatism occurs in nature, but it has not been possible until now to generate laboratory MFCI explosions from the interaction between high‐silica melts and water under laboratory conditions. The high viscosity of these melts apparently prevents formation of an effective hydrodynamic premix of melt and water, the documented precursor of experimental explosive MFCI caused by mafic melts. Our new experiments utilized samples from a rhyolitic tuff ring volcano in Mexico (Tepexitl). An experimental approach was developed, in which premixing conditions were generated by mechanical deformation of the melt, leading to brittle‐type fragmentation at the melt‐water interface. Physical measurements recorded during laboratory explosion provide quantitative evidence for rhyolitic explosive MFCI. Additionally, a comparison of experimentally produced particles with natural ones from Tepexitl deposits show nearly identical chemical/mineralogical composition, grain size, and grain morphology. Detailed textural analysis confirmed the presence of phreatomagmatically produced particles in both experimental and natural analog particles. The results from this series of experiments indicate that under natural conditions, stress‐induced magma fracturing can lead to a critical magma‐water‐interface growths and trigger phreatomagmatic explosions of high‐silica magma. The water source for these eruptions may include shallow aquifers, surface water bodies, strong precipitation, and intrusion into ice or wet, unconsolidated sediments.

Experimental investigation on the moving characteristics of molten metal droplets impacting coolant

This paper reports the results of an experimental investigation on the moving characteristics of molten metal droplets impacting coolant free surface. A visualization experimental facility of molten fuel coolant interactions (MFCI) is designed and set up in the present study. The lead–bismuth (Pb–Bi) alloys are employed as the metal materials. An automatic control circuit is designed and applied to control the release of the molten droplets. High-speed camera is employed to record the movement of the molten metal droplets falling down and into a coolant pool. Based on the analysis of the experimental data, a so-called “J-region” of the droplet’s velocity–time curves was found and the melt droplet enters the “J-region” when it impacts the free water surface. In the “J-region”, the velocity of the melt droplet decreases rapidly and suddenly at first, and then increases again. The droplet gradually reaches a comparatively steady velocity when it leaves the “J-region”. The present study provides essential information for further study on the fragmentation behavior of high-temperature molten droplets in coolant.

MFCI experiments on the influence of NaCl-saturated water on phreatomagmatic explosions

Molten–Fuel–Coolant Interaction (MFCI) experiments were performed using remelted foiditic rock samples from the West Eifel volcanic field (Germany). Two experimental series were carried out with one magmatic melt and two water compositions. Bi-distilled water was used in the first series (DW-1 to DW-5). In the second series (SW-1 to SW-5), the bi-distilled water was saturated (350 g L − 1 ) with sodium chloride (NaCl). For both experimental series the fragmentation history and the energy release were recorded and compared. The smallest particles (≤ 125 μm) were studied using scanning electron microscopy (SEM). Most MFCI experiments with bi-distilled water reached higher explosion intensities than the experiments with the saline water. This was accompanied by higher particle ejection velocities as well as the formation of more fine-grained and more interactive particles of angular shape. Additionally, the smallest artificial pyroclasts were examined by evolved gas analyses (EGA). The particles from the MFCI experiments with salt solutions are found to contain more sodium hydroxide (NaOH). These observations can be explained by thermodynamic arguments. In contrast to the MFCI experiments with pure water, an additional reaction occurs with saline water that results in evolution of hydrogen chloride (HCl) gas and leaves a residue of sodium hydroxide. The MFCI process with saline water consumes more enthalpy and Gibbs free energy, so that less energy is available for the explosion. With other sodium halides dissolved in the water (NaF, NaBr or NaI) the additional reaction can be predicted to have greater or lesser effects on phreatomagmatic explosions.

Peperite formed by lava flows over sediments: An example from the central Paraná Continental Flood Basalts, Brazil

Pahoehoe flows interbedded with sediments have been identified in the superior portion of Paraná Continental Flood Basalts (PCFB), west portion of Paraná State, southern Brazil. In the study area peperites are generated by the interaction between lava flows and wet lacustrine sediments (silt and clay). Evidence that the sediments were unconsolidated or poorly consolidated and wet when the lava flowed over them includes vesiculated sediment, sediment in vesicles and fractures in lava flow and in juvenile clasts in the peperite and soft sediment deformation. Hydrodynamic mingling of lava and wet sediments (coarse mingling) is predominant and volcanic rocks and textures related to explosive phase of Molten Fuel Coolant Interaction (MFCI) are not observed in study area. Locally centimeter-sized areas display direct contact between ash-sized juvenile clasts and sediments formed by the collapse of a vapor film. The textures of fluidal peperites in the central PCFB indicate that the relevant factors that led to a coarse mingling between lava/sediment are (1) lava properties (low viscosity); (2) fine grained, unconsolidated or poorly consolidated wet sediment; and (3) a single episode of interaction between lava flows and sediment.

ICONE11-36303 NUMERICAL INVESTIGATION ON THE BEHAVIOR OF A DROPLET AND VAPOR FILM DUE TO SHOCK WAVE

The droplet behavior in coolant is one of the key phenomena that govern the dynamics and energetics of molten fuel-coolant interactions (MFCI) during a postulated severe accident in a nuclear reactor. The violent interaction between droplet and coolant may lead to a steam explosion, which may contribute early failure of reactor vessel or containment. Phenomenology of steam explosion is very complex, involving intense thermal-hydrodynamics. With the high temperature melt droplet falling into the volatile liquid coolant, the droplet surface experiences intensive film boiling. The formation of vapor film on the droplet surface mitigates the deformation of the droplet because of the distinct properties. In one of hypothetical mechanisms of steam explosion, for example, it is believed that the collapse of the vapor film, triggered by a shock wave, enables liquid-liquid contact and potential for coolant to penetrate into the melt droplet. This phenomenon may cause excessive steam production and droplet fragmentation. Interest in modelling of MFCI has motivated experimental and analytical investigation on steam explosion. But steam explosion is so fast and complex phenomenon, it is very difficult (if not impossible) to capture every detail of the process. The recent development of computational fluid dynamics may improve significantly the understanding of the mechanism of the steam explosion during MFCI. In our previous works, an advanced front-capturing algorithm, called the level set method, was successfully developed for the simulation of two-phase (bubble or droplet) and three-phase (droplet with vapor film in coolant) flows. The interfaces were implicitly captured by smooth level set functions. This method leaves only one set of conservation equations to be solved for all the multi phases. A high order solver, named cubic-interpolated pseudo-particle (CIP) method, is used to solve conservation equations. In the present work, a shock wave is introduced to the multiphase system. For the purpose of comparison, the revolutions of both single bubble and single droplet due to shock wave are simulated, as well as behavior of a droplet covered by a vapor film under shock wave. Due to the complexity of the problem, only hydrodynamic process under isothermal conditions is studied. Under isothermal condition, the vapor film is introduced initially. A two-dimensional model was chosen. The shock wave is introduced at the bottom as the case in most single droplet steam explosion experiment. The other boundaries are set to Neumann (open) conditions. Through the simulation, further understanding on the mechanism of melt droplet evolution due to shock wave is obtained. Due to the high viscosity and density of droplet, the effect of shock wave on droplet deformation is less significant than that on bubble deformation. However, for the droplet with vapor film in a water pool, existence of the vapor film mitigates the droplet deformation properties due to shock wave. The collapse of vapor film and the penetration of water into the droplet are observed under certain conditions.

Thermohydraulic explosions in phreatomagmatic eruptions as evidenced by the comparison between pyroclasts and products from Molten Fuel Coolant Interaction experiments

Thermohydraulic explosions were produced by Molten Fuel Coolant Interaction (MFCI) experiments using remelted shoshonitic rocks from Vulcano (Italy). The fragmentation history and energy release were recorded. The resulting products were recovered and analyzed with the scanning electron microscope. Fine particles from experiments show shape and surface features that result from melt fragmentation in brittle mode. These clasts relate to the thermohydraulic phase of the MFCI, where most of the mechanical energy is released; they are here called “active” particles. The total surface area of such particles is proportional to the energy of the respective explosions. Other particles from experiments show shape and surface features that result from melt fragmentation in a ductile regime. These fragments, called “passive” particles, form after the thermohydraulic phase, during the expansion phase of the MFCI. In order to verify thermohydraulic explosions in volcanic eruptions, we compared experimental products with samples from phreatomagmatic base‐surge deposits of Vulcano. Ash particles from the experiments show features similar to those from the deposits, suggesting that the experiments reproduced the same fragmentation dynamics. To achieve discrimination between active and passive particles, we calculated shape parameters from image analysis. The mass of active particles in base‐surge deposits was calculated. As the material properties for the natural samples are identical to the experimental ones, the energy measurements and calculations of the experiments can be applied. For a single phreatomagmatic eruption at Vulcano, a maximum mechanical energy release of 2.75 × 1013 J was calculated, representing a TNT analogue of 6.5 kt.

Premixing of magma and water in MFCI experiments

Experimental studies have been performed to evaluate pre-explosive water–melt mixes with respect to explosive volcanic molten–fuel–coolant interaction (MFCI), i.e., phreatomagmatic explosion. Remolten ultrabasic volcanic rock was used as a magma simulant. Measurement of the explosion intensity was used to determine optimal premixing conditions. A well-defined optimal range was found for the hydrodynamic mixing energy (differential flow speed of 4.2 m/s), as well as for the water/melt mass ratio (0.03 to 0.04) under experimental conditions. The mass flux of water had a minor influence on the explosion intensity. Additionally, transparent mixing experiments with silicon oil and inked water were carried out. They indicate a direct dependence of the pre-explosive water-melt interface area on the explosion intensity. The experimental results show that the contact conditions of water and melt required for explosive MFCI may easily be established in natural volcanic systems. Thus, explosive MFCI is a probable mechanism of explosive volcanism.

Fragmentation of basaltic melt in the course of explosive volcanism

With the aim to enhance interpretation of fragmentation mechanisms during explosive volcanism from size and shape characteristics of pyroclasts experimental studies have been conducted using remelted volcanic rock (olivine‐melilitite). The melt was fragmented and ejected from a crucible by the controlled release of pressurized air volumes (method 1) or by controlled generation of phreatomagmatic explosions (Molten Fuel Coolant Interaction (MFCI); method 2). Both methods were adjusted so that the ejection history of the melt was identical in both cases. The experiments demonstrate that exclusively during MFCI, angular particles in the grain size interval 32 to 130 μm are generated that show surface textures dominated by cracks and pitting. The physical process of their generation is described as a brittle process acting at cooling rates of >106 K/s, at stress rates well above 3 GPa/m2, and during ∼700 μs. In this time period the emission of intense shock waves in the megahertz range was detected, releasing kinetic energy of >1000 J. By both experimental methods, three more types of particles were produced in addition, which could be identified and related to the acceleration and ejection history of the melt: spherical particles, elongated particles, and Pele's hair. Abundance and grain size distribution of these particles were found to be proportional to the rate of acceleration and the speed of ejection but were not influenced by the experimental method used. Pele's hair occurred at ejection speeds of >75 m/s.

Experimental study on simulated molten jet-coolant interactions

Molten fuel-coolant interactions in a jet contact mode was studied with respect to the safety of liquid-metal-cooled fast breeder reactors (LMFBRs). From a series of molten Wood's metal (melting point 79 °C) jet-water interaction experiments, several distinct modes of interaction behaviors were observed for various combinations of initial temperature conditions of the two fluids. A semi-empirical model for a minimum film boiling temperature criterion was developed and used to reasonably explain the different interaction modes. It was concluded that energetic jet-water interactions are only possible under relatively narrow initial thermal conditions. Preliminary extrapolation of the present results to a UO2-sodium system suggests that mild interactions and coolable debris formation should be most likely in LMFBRs.