Metal Spallation Research Articles

Preliminary safety analysis of the TRANSMUTEX sub-critical reactor using the GeN-Foam multi-physics solver

This work presents a preliminary safety analysis of the accelerator-driven system under development at TRANSMUTEX. It is a sub-critical lead-cooled fast reactor with a nominal power of 300 MWth having the following features: (1) an inert-matrix fuel; (2) a keff = 0.974; (3) a heavy metal spallation target; and (4) a 4 MW proton accelerator. The safety analyses are based on the GeN-Foam multi-physics code, whose point-kinetics sub-solver has been extended for the analysis of sub-critical reactor configurations. In the paper, we first present the extension of the code and its verification against analytical results. Then, we present the results of our preliminary safety analysis, including in particular the simulation of various unprotected (double-fault) accidental transients.

Multi-scale simulation study on characteristics of free surface velocity curve in ductile metal spallation

The spallation characteristics of ductile tantalum metal under planar plate impact was analyzed through a multi-scale perspective. And the typical characteristics of the free-surface velocity curve on the macro-scale were interpreted from the micro-scale to reveal the physical meanings corresponding to these typical characteristics. On the macro-scale, the spallation behaviors of the ductile tantalum metal under planar-plate impact were numerically simulated through the smooth particle hydrodynamics (SPH) and Lagrange methods, and the free-surface velocity curves of the tantalum during spallation were obtained. In addition, the free-surface velocity curves obtained by the Johnson-Cook model, Steinberg-Cochran-Guinan model and Zerilli-Armstrong model were compared in the numerical simulations. Comparison with the experimental data shows that the Steinberg-Cochran-Guinan constitutive model has a better performance in the macro-level simulation. The free-surface velocity curves at different strain rates were obtained by changing the loading conditions, and the typical characteristics of the free-surface velocity curves at different strain rates were discussed. Results show that there is an exponential relationship between spall strength and strain rate, and the spall strength obtained from the simulation has a good agreement with the experimental data. On the micro-scale, the damage evolution in the spallation region was obtained by molecular dynamics simulation conducted in the LAMMPS software, and the loading strain rate was consistent with that on the macro-scale. The micro-scale simulation reveals the physical connotation of the typical characteristics of the macro-scale free-surface velocity curve. Micro-scale analysis shows that spallation is the response of damage evolution of nucleation, growth, and aggregation of voids. From the multi-scale perspective analysis, the typical characteristics on the free-surface velocity curve are closely related to the damage evolution in the spallation area: the pullback signal is a macroscopic response of the void nucleation in the spall area; the decline amplitude of the free-surface velocity curve reflects the void nucleation condition, and the spall strength reflects the nucleation strength of the voids. What’s more, the velocity rises to the first peak beyond the minima after the pullback signal reflects the rate of damage evolution. The multi-scale perspective analysis is helpful to fully understand the physical mechanism of the spallation under planar-plate impact.

Cavitation-induced erosion in liquid metal spallation targets – A comparison between a predictive method and measured erosion damage to stainless steel target modules at the Spallation Neutron Source

Neutrons are produced for neutron scattering instruments at the Spallation Neutron Source by bombarding liquid mercury with a pulsed high-energy proton beam. The liquid mercury target material flows through a replaceable stainless-steel target vessel and cavitates during operation due to the extremely high heating rate caused by each proton pulse. Cavitation-induced erosion damage to the mercury-facing surfaces of target vessels is a key concern for target lifetime and has been the subject of considerable research. While numerous experiments have studied and modeled the progression of cavitation erosion damage to liquid metal target containers, direct quantitative measurements of actual erosion of a spallation target vessel were previously not available. Recently, the erosion damage to multiple target vessels operated at the Spallation Neutron Source was measured using a high-resolution laser scanning system, which provided an opportunity to study the efficacy of a cavitation damage progression model. In this paper, a cavitation progression modeling procedure developed using empirical observations is compared to measurements of samples from target vessels after operation. The strengths and limitations of the technique are discussed and methods for improving future modeling and predictive accuracy are proposed.

Assessment of the power deposition on the MEGAPIE spallation target using the GEANT4 toolkit

This work aims at evaluating the reliability of the GEANT4 (GEometry ANd Tracking 4) Monte Carlo (MC) toolkit in calculating the power deposition on the Megawatt Pilot Experiment (MEGAPIE), the first liquid–metal spallation target worldwide. A new choice of codes to study and optimize this target is provided. The evaluation of the GEANT4 toolkit is carried out in comparison with the MCNPX and FLUKA MC codes. The MEGAPIE is an international project led by the Paul Scherrer Institute in Switzerland. It aims to demonstrate the safe operation of an intense neutron source to power the next generation of nuclear reactors, accelerator-driven systems (ADSs). In this study, we used the GEANT4 MC toolkit to calculate the power deposited by fast protons on the MEGAPIE target. The calculation focuses on several structures and regions. The predictions of our calculations were compared and discussed with that of the MCNPX and FLUKA codes, adopted by the MEGAPIE project. The comparison shows that there is a very good agreement between our results and those of the reference codes.

Three-dimensional CFD simulations to study the effect of impeller geometry on internal flow field in ADS upward spiral flow target

ABSTRACT The relationship between flow field characteristics and geometrical structures is an important issue in designing liquid metal spallation target for accelerator-driven sub-critical system (ADS). Upward spiral flow target (USFT) is a new type ADS windowless target in which an impeller is employed to generate upward spiral flow and regulate internal flow field. The impeller comprises a central shaft and several identical blades, and its geometry is very crucial for designing the target. In this paper, series models with different geometric parameters of the shaft and the blades are simulated by ANSYS CFX to analyse the distributions of internal flow field and the dimensions of stagnation zone in USFT. The results show that the shaft plays a primary role in guiding the upward flow while the blades are significant for generating the spiral flow. Finally, an optimized model with preferred geometric parameters is obtained by comparing the overall simulation results. The whole research has foundational signification for further studying USFT.

Preparation and verification of mixed high-energy neutron cross-section library for ADS

An accelerator-driven subcritical system (ADS) is driven by an external spallation neutron source, which is generated from a heavy metal spallation target to maintain stable operation of the subcritical core, where the energy of the spallation neutrons can reach several hundred mega-electron volts. However, the upper neutron energy limit of nuclear cross-section databases, which are widely used in critical reactor physics calculations, is generally 20 MeV. This is not suitable for simulating the transport of high-energy spallation neutrons in the ADS. We combine the Japanese JENDL-4.0/HE high-energy evaluation database and the ADS-HE and ADS 2.0 libraries from the International Atomic Energy Agency and process all the data files for nuclides with energies greater than 20 MeV. We use the continuous pointwise cross-section program NJOY2016 to generate the ACE-formatted cross-section data library IMPC-ADS at multiple temperature points. Using the IMPC-ADS library, we calculate 10 critical benchmarks of the International Criticality Safety Benchmark Evaluation Project manual, the 14-MeV fixed-source problem of the Godiva sphere, and the neutron flux of the ADS subcritical core by MCNPX. To verify the correctness of the IMPC-ADS, the results were compared with those calculated using the ENDF/B-VII.0 library. The results showed that the IMPC-ADS is reliable in effective multiplication factor and neutron flux calculations, and it can be applied to physical analysis of the ADS subcritical reactor core.

Fourier amplitude sensitivity test applied to dynamic combined finite‐discrete element methods–based simulations

SummaryFracture propagation plays a key role for a number of applications of interest to the scientific community, from dynamic fracture processes like spallation and fragmentation in metals to failure of ceramics, airplane wings, etc. Simulations of material deformation and fracture propagation rely on accurate knowledge of material characteristics such as material strength and the amount of energy being dissipated during the fracture process. Within the combined finite‐discrete element method (FDEM) framework material fracture behavior is typically described through a parametrized softening curve, which defines a stress‐strain relationship unique to each material. We apply the Fourier amplitude sensitivity test to explore how each of these parameters influences the simulated damage processes and to determine the key input parameters that have the most impact on the model response. We present several sensitivity numerical experiments for the simulation of a split Hopkinson pressure bar (SHPB) test for weathered granite samples using different combinations of model parameters. We validate the obtained results against SHPB experimental data. The experiments show that the model is mostly sensitive to parameters related to tensile and shear strengths, even in the presence of other parameter perturbations. The results suggest that the specification of tensile and shear strengths at the interfaces dominate the stress‐time history of the FDEM simulation of SHPB test.

Detection of the multiple spallation parameters and the internal structure of a particle cloud during shock-wave loading of a metal

The results of experiments on studying spallation and the ejection of particles from the surfaces of copper and lead samples are presented. A laser interferometry method is used to detect the particle cloud velocity and the multiple spallation parameters. Angular detectors are used to detect the depth profile of the particle cloud velocity dispersion and the structure of metal spallation.

Radiochemical determination of 129I and 36Cl in MEGAPIE, a proton irradiated lead-bismuth eutectic spallation target

Abstract The concentrations of the long-lived nuclear reaction products 129I and 36Cl have been measured in samples from the MEGAPIE liquid metal spallation target. Samples from the bulk target material (lead-bismuth eutectic, LBE), from the interface of the metal free surface with the cover gas, from LBE/steel interfaces and from noble metal absorber foils installed in the cover gas system were analysed using Accelerator Mass Spectrometry at the Laboratory of Ion beam Physics at ETH Zürich. The major part of 129I and 36Cl was found accumulated on the interfaces, particularly at the interface of LBE and the steel walls of the target container, while bulk LBE samples contain only a minor fraction of these nuclides. Both nuclides were also detected on the absorber foils to a certain extent (≪ 1% of the total amount). The latter number is negligible concerning the radio-hazard of the irradiated target material; however it indicates a certain affinity of the absorber foils for halogens, thus proving the principle of using noble metal foils for catching these volatile radionuclides. The total amounts of 129I and 36Cl in the target were estimated from the analytical data by averaging within the different groups of samples and summing up these averages over the total target. This estimation could account for about half of the amount of 129I and 36Cl predicted to be produced using nuclear physics modelling codes for both nuclides. The significance of the results and the associated uncertainties are discussed.

Radiochemical Determination of Polonium in Liquid Metal Spallation Targets

The MEGAPIE target, consisting of 82 litres of lead-bismuth eutectic (LBE), was irradiated close to the megawatt range (0.8 MW) from August to December 2006 in the SINQ facility at PSI. After a cooling period of 5 years, a post-irradiation examination (PIE) program was started and samples were taken from different positions in the target. In this paper we focus on the measurement of α-emitting 208-210Po in the MEGAPIE target. The experimental results are compared with theoretical predictions obtained by FLUKA and MCNPX calculations.

Analysis of the 207Bi, 194Hg/Au and 173Lu distribution in the irradiated MEGAPIE target

Samples obtained from different locations within the prototype liquid metal spallation target MEGAPIE irradiated in 2006 at PSI were analysed using γ-spectrometry. A variety of radionuclides formed by reaction of the target material, lead–bismuth eutectic (LBE), with the proton beam and secondary particles were identified. While nuclides representing the target material itself (207Bi) and nuclides of noble metals were found in LBE samples throughout the target, nuclides of electropositive metals were found to be quantitatively deposited on free surfaces and material interfaces within the target system. This behaviour is analysed in more detail based on results obtained for three nuclides representing groups of elements with distinct chemical behaviour, namely 207Bi, 194Hg/Au and 173Lu. Quantitative analysis results are given and compared with predictions obtained using nuclear physics calculations for those nuclides showing rather homogeneous distribution within the target. Possible reasons for the separation of radionuclides from the liquid metal and their deposition on surfaces are given, and consequences arising for nuclear facilities utilizing liquid metals are discussed.

Simulation of impact/explosive driven spallation in metals: comparative study of damage and dynamic strength models

Spallation refers to fracture occurring in materials due to tensile loading. This paper presents an overview of our extensive one- and two-dimensional simulations done to study spallation in impact/explosive loaded Copper or Mild Steel targets, using an Arbitrary-Lagrangian-Eulerian hydrocode. Three methods of computing the spall strength and spall thickness have been employed. The computed spall parameters have been compared with Russian spall experiments. In impact loaded targets, due to a square shock wave, only one scab has been observed. In the case of explosive loading, as the rarefaction associated with the incoming shock wave is triangular, there occur many high-tension regions leading to the formation of multiple scabs (typically 2-3 scabs). Effect of flier velocity and target temperature on spallation has been studied. These trends match with experiments. Edge effects due to finite diameter of the target attenuate the incoming shock wave by lateral release, resulting in more marked damage near the axis than on the periphery. Four damage models and three dynamic strength models have been examined to determine the best-suited model for spallation studies. It is found that the Void Growth (VG) damage model and Zerilli-Armstrong strength model yield spall parameters close to experiments.

Initial observations of cavitation-induced erosion of liquid metal spallation target vessels at the Spallation Neutron Source

During operation of the Spallation Neutron Source (SNS) at Oak Ridge National Laboratory the mechanical properties of the AISI 316L target module are altered by high-energy neutron and proton radiation. The interior surfaces of the target vessel are also damaged by cavitation-induced erosion, which results from repetitive rapid heating of the liquid mercury by high-energy proton beam pulses. Until recently no observations of cavitation-induced erosion were possible for conditions fully prototypic to the SNS. Post-Irradiation Examination (PIE) of the first and second operational SNS targets was performed to gain insight into the radiation-induced changes in mechanical properties of the 316L target material and the extent of cavitation-induced erosion to the mercury vessel inner surfaces. Observations of cavitation-induced erosion of the first and second operational SNS target modules are presented here, including images of the target vessel interiors and specimens removed from the target beam-entrance regions.

Radiochemical aspects of liquid mercury spallation targets

Liquid metal spallation targets using mercury as target material are used in state-of-the-art high power pulsed neutron sources that have been constructed in the USA and Japan within the last decade. Similar target concepts were also proposed for next generation ISOL, β-beam and neutrino facilities. A large amount of radioactivity will be induced in the liquid metal during operation caused by the interaction of the target material with the intense proton beam. This radioactivity – carried by a wide range of radioisotopes of all the elements of the periodic table from hydrogen up to thallium – must be considered for the assessment of safe operation and maintenance procedures as well as for a final disposal of the used target material and components. This report presents an overview on chemical investigations performed in our laboratory that deal with the behavior of radionuclides in proton irradiated mercury samples. The solubility of elements in mercury was calculated using thermodynamical data obtained by a semi-empirical method. It is found that several important nuclear reaction products will reach their solubility limit in a high power mercury spallation target. It is also experimentally proven that the radionuclides produced by proton irradiation are generally not homogeneously distributed in the liquid metal but tend to accumulate on vessel walls and free surfaces. The consequences resulting from this behavior for the operation of mercury spallation targets are discussed.

Vapour phase concentrations of volatile nuclear reaction products in the MEGAPIE cover gas

The licensing of the prototype liquid lead bismuth spallation target MEGAPIE required the assessment of the amount of radioactivity that can be released from the liquid metal to the gas phase under various scenarios. To estimate the radiological consequences of evaporation processes under normal operation conditions, the concentrations of hazardous volatile radionuclides in the gas phase of the MEGAPIE expansion volume were estimated using a simplified model based on an equilibrium state. In this report, we focus on those volatile impurities that do not show a strong retention in the liquid metal caused by chemical interactions, comprising mercury, cadmium and thallium. For mercury, temperature functions for the effective vapour pressure of mercury over liquid eutectic lead bismuth alloy were deduced from experimental data and compared to literature data available for the binary systems Hg–Pb and Hg–Bi. Conservative functions were selected from the data on hand for the evaluation of the maximum possible amounts of mercury, cadmium and thallium radioactivity in the gas phase of the MEGAPIE target. Substantial amounts of radioactive mercury are predicted to be released to the cover gas phase of MEGAPIE and liquid lead-alloy based spallation targets in general. The radioactivity resulting from evaporation of cadmium and thallium nuclides is expected to be low. Consequences for liquid metal spallation target systems are discussed.

Mesoscale polycrystal calculations of damage in spallation in metals

The goal of this project is to produce a damage model forspallation in metals informed by the polycrystalline grain structure at themesoscale. Earlier damage mod- els addressed the continuum macroscale in whichthese effects were averaged out. In this work we focus on cross sectionsfrom recovered samples examined with EBSD (elec- tron backscattereddiffraction), which reveal crystal grain orientations and voids. We seek tounderstand the loading histories of specific sample regions by meshing upthe crystal grain structure of these regions and simulating the stress,strain, and damage histories in our hydrocode, FLAG. The stresses and strainhistories are the fundamental drivers of damage and must be calculated. Thecalculated final damage structures are compared with those from therecovered samples to validate the simulations.

Effect of tensile stress on cavitation damage formation in mercury

Cavitation erosion or so called pitting damage was investigated under tensile stress conditions in mercury. In MW-class liquid metal spallation targets, pitting damage is a critical issue to satisfy required power and/or lifetime of the target vessel. Cavitation occurs by negative pressure which is induced through pressure wave propagation due to proton beam injection. Pitting damage is formed by microjet and/or shock wave during cavitation bubble collapse. A mercury target vessel suffers tensile stress due to thermal stress or welding. In order to investigate the effect of tensile stress on pitting damage formation, cavitation erosion tests were performed using stress imposed specimens in mercury. An ultrasonic vibratory horn and electro-Magnetic IMpact Testing Machine (MIMTM) were used to vary the cavitation intensity. In the incubation period of pitting damage, damaged area was slightly increased with increasing imposed tensile stress. In the steady state period, a mean depth of erosion was increased by the tensile stress. Additionally, in order to quantitatively evaluate the effect of tensile stress, an indentation test with Vickers indenter was carried out to quasi-statically simulate the impact load. From the measurement of the diagonal length of the indent aspect ratio and hardness, it is recognized that the threshold of the deformation, i.e. pitting damage formation, was decreased by the tensile stress.

Analysis of corrosion scale structure of pre-oxidized stainless steel 316 in molten lead bismuth eutectic and the relation to impedance spectroscopy response

Stainless steel 316 is a candidate structural material for heavy liquid metal spallation neutron sources and fast reactor systems. Impedance spectroscopy (IS) provides a means of real-time measurements of the stability of protective oxide scales in these systems. The IS technique has been shown to yield quantitative results at lower temperatures, but does not yet yield quantitative results at temperatures of interest for engineering applications. To understand this, pre-oxidized stainless steel 316 samples were immersed in oxygen-saturated stagnant lead bismuth eutectic at 550 °C for 1335 h. During the exposure, real-time impedance spectroscopy was taken from different samples. Negligible impedance was observed from one sample at the end of 1335 h, indicating that the oxide scale was non-protective. A variety of microanalysis techniques revealed that oxide layers formed on the sample surface suffered corrosion attack from the LBE. A duplex oxide structure composed of outer iron oxide and inner iron chromium oxide was found on some sample surface area. The major phases of the duplex oxide scales were Fe 3O 4 and FeCr 2O 4. The high conductivity of this duplex scale explains the corrosion rate as well as the lack of IS response.

Post-irradiation examination of the Spallation Neutron Source target module

The Spallation Neutron Source (SNS) at Oak Ridge National Laboratory is an accelerator-based pulsed neutron source that produces high-energy spallation neutrons by bombarding liquid mercury flowing through a stainless steel target vessel. During operation the proton beam and spallation neutrons produce radiation damage in the AISI 316L austenitic stainless steel target vessel and water-cooled shroud. The beam pulses also cause rapid heating of the liquid mercury, which may produce cavitation erosion damage on the inner surface of the target vessel. The cavitation erosion rate is thought to be highly sensitive to beam power and predicted to be the primary life-limiting factor of the target module. Though cavitation erosion and radiation damage to the target vessel are expected to dictate its lifetime, the effects of radiation damage and cavitation erosion to target vessels in liquid metal spallation systems are not well known. Therefore preparations are being undertaken to perform post-irradiation examination (PIE) of the liquid mercury target vessel and water-cooled shroud after end-of-life occurs. An overview of the planned PIE for the SNS target vessel is presented here, including proposed techniques for specimen acquisition and subsequent material properties characterization.

Neutronic characterization of the MEGAPIE target

The MEGAPIE project aimed to design, build and operate a liquid metal spallation neutron target of about 1 MW beam power in the SINQ facility at the Paul Scherrer Institut (Villigen, Switzerland). This project is an important step in the roadmap towards the demonstration of the accelerator driven system (ADS) concept and high power liquid metal targets in general. Following the design phase, an experimental program was defined to provide a complete characterization of the facility by performing a “mapping” of the neutron flux at different points, from the center of the target to the beam lines. The neutronic performance of the target was studied using different experimental techniques with the goals of validating the Monte Carlo codes used in the design of the target; additionally, the performance was compared with the solid lead targets used before and after the MEGAPIE experiment.