Boiling Water Reactor Research Articles

Two-group bubble interfacial area concentration correlation for upward bubbly to churn flows in vertical rod bundle flow channels

The detailed understanding and accurate modeling of the upward two-phase flows in vertical rod bundle flow channels are significant for various industrial applications, such as performing the optimal design and evaluating the transient and accident scenarios for BWR (Boiling Water Reactor) and PWR (Pressurized Water Reactor) systems. The interfacial area concentration (IAC) describing the available interfacial area for the transfer of mass, momentum, and energy between two phases plays an important role in the two-phase flow simulation in the frequently-used thermal-hydraulics system codes. Although various IAC correlations utilizing the drag-based two-group bubble categorization, namely two-group bubble IAC correlations, have been proposed to predict the IAC changes in the two-phase flow ranging from bubbly to churn flows in simple-geometry flow channels, the two-group bubble IAC correlation for rod bundle flow channels is still underdeveloped. This study performed a survey on the existing representative two-group bubble IAC correlations for simple-geometry flow channels and collected 425 experimental IAC data taken in various vertical rod bundle flow channels. The predictive abilities of these two-group bubble IAC correlations were evaluated with the collected experimental IAC data. The evaluations showed that no existing correlations can consistently maintain satisfied predictive ability throughout the upward bubbly to churn flows in vertical rod bundle flow channels. In view of this fact, this study proposed a new two-group IAC correlation for vertical rod bundle flow channels. Specifically, the group-2 bubble void fraction model was proposed by physically analyzing the changing characteristics of the group-2 bubbles in different flow regimes. The Sauter mean diameter model of group-1 bubbles was derived by (a) selecting the distorted small bubble as the representative group-1 bubble for the bubbly, cap-bubbly and churn flows and (b) selecting the spherical small bubble as the representative group-1 bubble for the finely dispersed bubbly flow. The Sauter mean diameter model of group-2 bubbles was proposed by treating the cap-bubble as the representative group-2 bubble and taking into account the bubble deformation caused by the surrounding rods. The proposed two-group bubble IAC correlation has been checked against the collected existing experimental IAC data and the proposed correlation exhibited reliable predictive abilities in the bubbly to churn flows in various rod bundle flow channels with the mean relative deviation of 0.233, 0.212, 0.247, 0.222 and 0.234 for bubbly, finely dispersed bubbly, cap-bubbly, churn and all data, respectively.

Model for determining rupture area in Zircaloy cladding under LOCA conditions

The dispersal of nuclear fuel through a cladding rupture opening during a transient event is the primary safety concern preventing the United States nuclear industry from extending burnup beyond a peak rod average burnup of 62 GWd/tU. The majority of research thus far has prioritized understanding, or at a minimum prioritized quantifying, fuel fragmentation and pulverization as a function of burnup and to a lesser extent onset temperature. Early high burnup loss-of-coolant accident tests performed at the Halden Boiling Water Reactor evaluated the impact of relocation on cladding temperature, whereas historical and recent cladding burst tests focused primarily on generating data related to cladding balloon and burst timing. However, very little effort has been put forth to develop a relationship for determining the rupture opening dimensions for Zircaloy cladding tubes bursting during a simulated light water reactor loss-of-coolant accident. This is significant because high burnup loss-of-coolant accident data suggest dispersal could be limited by the size of the rupture opening or rupture dimensions, thereby minimizing the safety concern. Therefore, the purpose of this research is to build a database composed of all simulated loss-of-coolant accident data to develop an empirical model that conservatively captures the rupture opening dimensions and translates those dimensions to rupture area. Analysis of the experimental data indicates the rupture dimensions are strongly correlated to the peak balloon strain. This relationship is leveraged to develop a rupture length empirical correlation as a function of peak balloon strain and a rupture width empirical correlation as a function of rupture length. Outliers in the experimental data are identified and discussed. Finally, the rupture opening empirical model is presented in such a way for general implementation into high fidelity fuel performance codes or fuel performance codes used for safety analyses.

MOOGLE: A Multi-Objective Optimization tool for three-dimensional nuclear fuel assembly design

MOOGLE is a new genetic algorithm based methodology for the 3D design of nuclear fuel assemblies. MOOGLE uses common fuel rod types as the decision variable to develop a suite of 3D fuel assemblies to provide optimized solutions to the design problem. Pressurized water reactor (PWR) fuel assemblies were optimized using Integral Fuel Burnable Absorber (IFBA) and gadolinium (Gd2O3) as burnable poisons to compare how burnable poison choice affects optimization results. Boiling water reactor (BWR) fuel bundles were also optimized using three unique fuel rod palettes to study how the size of the design space affects optimization results. Burnable poison analysis showed that utilizing IFBA and Gd2O3 as burnable poisons produced the best and widest range of optimized solutions. BWR fuel bundle optimization results indicate that the inclusion of additional fuel rod types produced a wider solution space but did not improve optimization results for regions explored using fewer unique fuel rods. These tests demonstrate MOOGLE’s ability to analyze the trade-offs between the inclusion of different fuel elements and their effects on assembly performance.

Effect of gas density and surface tension on liquid film thickness in vertical upward disturbance wave flow

Gas–liquid annular flows, such as the flow attached to the fuel rods of boiling water reactors, is a prevalent occurrence in industrial processes. At the gas–liquid interface of such flows, disturbance waves with diverse heights commonly arise. Because the thin liquid film between two successive disturbance waves leads to the dryout on the heating surface and limits the performance of the heating components, complete knowledge of the film thickness and disturbance waves is crucial. The properties of the base film and disturbance waves have been studied extensively through experimental and analytical approaches. However, most experimental data and analysis available in the literature are limited to the near atmospheric condition, and the liquid–gas density ratio under near atmospheric conditions is much higher than in practical nuclear applications. In this study, we use nitrogen gas and water under the system pressures of 0.2 and 0.4 MPa, as well as HFC134a gas and water under the system pressure of 0.7 MPa. Therefore, the density ratio can be varied between 32 and 434. Based on our experimental data, we report a direct link between the film thickness, wave height, and dimensionless numbers. The analysis indicates the Weber number which represents the ratio of the drag force to the surface tension force is critical for the film thickness behavior. To examine this, we perform the experiments using nitrogen gas and 95% (v/v) of aqueous ethanol solution under the system pressure of 0.2 MPa. Finally, two models are proposed to predict the averaged film thickness and wave height, which are then compared with previous data.

Recycling alternatives for minor actinide and plutonium in a boiling water reactor

Recycling is an alternative for reducing minor actinide and plutonium stockpiles from depleted fuel assemblies. In the current study, it is proposed the use of a MOX fuel assembly called MOXIM-MA. It comprises 4 rods with minor actinides, 14 UO2-gadolinium rods, 8 water rods, and 66 MOX fuel rods. This MOXIM-MA design is used in a 2027 MWth BWR, reducing 0.130 kg HM of minor actinide and 5.042 kg HM of Plutonium per fuel assembly at discharge after discharge 4-operating cycles of 18 months each. The reactor core energy produced does not suffer any penalty by using this fuel assembly. There is no need for additional changes to the original control safety system. The radiotoxicity results show almost one magnitude order increase compared with standard UO2 fuel assemblies for the same BWR reactor core. Still, this increase is the same magnitude experienced when MOX fuel assemblies without minor actinide are used. A twin reactor power plant, 2027 MWth per unit, with 20 and 25 years of operational life, is considered to propose a plutonium and minor actinide recycling strategy. Standard UO2 fuel assemblies, at discharge for this plant, produce 0.135 kg HM of MA and 1.707 kg HM of plutonium. By recycling them, at least 6 depleted fuel assemblies are needed for each MOXIM-MA fuel assembly processed. The suggested strategy starts by considering the cumulative amount of minor actinide and plutonium coming from the depleted fuel assemblies discharged during the plant's operating life at the moment of this analysis. The study shows that the MOXIM-MA fuel assemblies will be used in only one of the reactor units in a once-through scheme. It describes how the auto recycling of plutonium and MA could be done to supply the MOXIM-MA fuel assemblies during the rest of the plant's operation. Results promise to deliver the plant's whole operation life a decrease of 39.50% in plutonium inventories and 12.70% in MA inventories, achieving this goal without penalizing energy production.

Quantification of Deep Neural Network Prediction Uncertainties for VVUQ of Machine Learning Models

Recent performance breakthroughs in artificial intelligence (AI) and machine learning (ML), especially advances in deep learning, the availability of powerful and easy-to-use ML libraries (e.g., scikit-learn, TensorFlow, PyTorch), and increasing computational power, have led to unprecedented interest in AI/ML among nuclear engineers. For physics-based computational models, verification, validation, and uncertainty quantification (VVUQ) processes have been very widely investigated, and many methodologies have been developed. However, VVUQ of ML models has been relatively less studied, especially in nuclear engineering. This work focuses on uncertainty quantification (UQ) of ML models as a preliminary step of ML VVUQ, more specifically Deep Neural Networks (DNNs) because they are the most widely used supervised ML algorithm for both regression and classification tasks. This work aims at quantifying the prediction or approximation uncertainties of DNNs when they are used as surrogate models for expensive physical models. Three techniques for UQ of DNNs are compared, namely, Monte Carlo Dropout (MCD), Deep Ensembles (DE), and Bayesian Neural Networks (BNNs). Two nuclear engineering examples are used to benchmark these methods: (1) time-dependent fission gas release data using the Bison code and (2) void fraction simulation based on the Boiling Water Reactor Full-size Fine-Mesh Bundle Tests (BFBT) benchmark using the TRACE code. It is found that the three methods typically require different DNN architectures and hyperparameters to optimize their performance. The UQ results also depend on the amount of training data available and the nature of the data. Overall, all three methods can provide reasonable estimations of the approximation uncertainties. The uncertainties are generally smaller when the mean predictions are close to the test data while the BNN methods usually produce larger uncertainties than MCD and DE.

Iterative computed tomography reconstruction of void fraction in a fuel bundle on unstructured mesh

The experimental investigation of internal two-phase flow structures in high pressure/high temperature systems is technologically challenging and limited to a few instrumentation techniques. For complex fuel bundle geometries, the use of Computed Tomography (CT) allows for relatively high spatial resolution (0.2 to 1 mm) and non-intrusive measurement of time-averaged void fraction. One of the main issues using standard structured reconstruction mesh is due to the loss of accuracy near the test section boundaries, for instance within the pixels overlapping both the two-phase coolant and the heated rods. In this work, a void fraction reconstruction methodology based on unstructured mesh is developed to accurately bound relevant attenuation regions within the considered test section. This method eliminates any overlapping reconstruction pixels and allows straightforward applications of knowledge-based, low noise, algebraic iterative reconstruction techniques (ART, SIRT, etc.). Within the same consistent framework, CT scan forward projections can be simulated for various selected cases (single-phase, arbitrary-two-phase void fraction distributions, etc.) and reconstructed to assess the performance of both the CT scan setup (e.g., projection data sampling and number of projections) and the reconstruction technique. The method is applied to the void fraction reconstruction of gamma CT data in a 5 × 5 fuel bundle geometry measured at the Westinghouse FRIGG thermal–hydraulic test facility operating under prototypical conditions of Boiling Water Reactor. The unstructured mesh is designed to match the features of the test section (isolation, pressure vessel, water gap, fuel channel, heated rods and coolant) with high-resolution in the domain of interest (coolant) and very low-resolution in all other domains with known attenuation coefficients. As compared to previous void fraction reconstruction attempts for this data set, the new reconstruction methodology allows significant improvements in void fraction resolution and noise reduction.

Development of a multipoint liquid film sensor using an optical waveguide film on simulated fuel rod of BWRs

ABSTRACT Measurement techniques for distribution of liquid film thickness (LFT) with high spatial and high temporal resolutions are necessary for evaluation of thermal margins in boiling water reactors. To experimentally investigate high-speed phenomena of the liquid film on a simulated fuel rod, we developed an LFT sensor using an optical waveguide film (OWF) with 42 μm spatial resolution and 1 MHz time resolution. The OWF is flexibly embedded in the surface of the simulated fuel rod and includes multiple light paths. Micro-mirrors located at ends of the light paths function as sensors and intensity of light reflected at the liquid film surface is detected. First, we determined the relationship between the LFT and the light intensity in a static liquid film. Second, signal processing for the OWF was shown by simulating random waves to extract influences of the surface angle on the output signals. Finally, we demonstrated measurement of the LFT in vertical annular flows with two sensors of the OWF. Mean LFT measured by the OWF gave good agreement with results from visualization within −20 to +10%. Furthermore, evaluation of the wave velocities for disturbance waves and ripples showed them to correspond to the visualization results.

Corrosion behavior of a series of combinatorial physical vapor deposition coatings on SiC in a simulated boiling water reactor environment

The simulated boiling water condition corrosion properties of multilayer and combinatorial physical vapor deposition metal coatings (Zr, Cr, and/or Ti) deposited on a SiCf-SiC composite were investigated. Various compositions within the ternary system were corroded to the most suitable composition preventing underlying SiCf-SiC from further attack. Mass measurements and visual observations were conducted to identify which coatings formed protective oxides and which coatings spalled. Transmission electron microscopy was conducted to observe the microstructure of the physical vapor deposition metal layer before and after corrosion as well as identify the oxide which were forming.

Improvement of two-phase closure models in CTF using Bayesian inference

Under the Consortium for Advanced Simulation of Light Water Reactors (CASL) program, extensive capabilities have been developed in CTF to analyze light-water reactors (LWRs) for normal operating conditions, departure from nucleate boiling (DNB), and system transients. However, further improvements are required in the modeling and simulation of boiling water reactors (BWRs), which is a focus of the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. In this work, CTF validation results were used to optimize selected modeling coefficients by calibrating to experimental data using a Bayesian inference approach. Calibration studies were conducted to improve (vapor) void fraction prediction without worsening the two-phase pressure drop prediction, as well as to improve the two-phase pressure drop prediction. Calibration was performed for interfacial drag and wall shear models. Surrogates were developed to alleviate the computational expense required for sampling the parameter space using Markov chain Monte Carlo (MCMC). An assessment performed with calibrated models demonstrated an improvement of CTF in its prediction of key parameters such as void fraction and two-phase pressure drop.

Experimental Evaluation of Two-Phase Heat Transfer Coefficient Under Oscillatory Flow Conditions and Correlation Development

Abstract Natural circulation boiling water reactors (NCBWRs) are prone to flow instabilities. Due to instabilities, heat transfer deteriorates. However, the extent to which the heat transfer coefficient (HTC) deteriorates is not well established. In this work, experiments are conducted under controlled flow oscillations to evaluate the impact of the time period, mean mass flux, and amplitude ratio on HTC. Existing correlations for HTC are compared with experimental results and are found to be not satisfactory. To improve the predictions of HTC, correction factors (CFs) are proposed for existing correlations. Further, an improved HTC correlation has been developed, incorporating new nondimensional numbers.

The interface design of storage and examination system for BWR spent nuclear fuel dry storage demonstration project in Taiwan

Due to the lack of Boiling Water Reactor (BWR) fuel examination data after dry storage, Taiwan Power Company and Institute of Nuclear Energy Research jointly launched the “Taiwan Nuclear Fuel Storage Demonstration Project” to collect fuel characteristics data under simulated dry storage. To meet the project requirements, BWR fuel rods in different fuel designs will be selected and then store in simulated storage containers in different environments for 10 years. The interface considerations and design results of the auxiliary equipment for spent nuclear fuel storage, transfer, and non-destructive testing in the simulation stage of spent fuel dry storage were summarized. In addition, the possibility of operating at high heating power and avoiding deformation of the components due to thermal extension of the storage containers after heating was included in the design consideration. A simplification concept was introduced, with frequent NDT (Non-destructive Testing) inspections that will be performed through the spent fuel transfer process. The test results showed that the interface design can meet the requirements of long-term experiment, maintenance and radiation protection requirements, and is ready for further demonstration.

Verification and validation of the Alternative Nonlinear Two-phase Subchannel (ANTS) code

The Alternative Nonlinear Two-phase Subchannel solver (ANTS) code was written to provide a fast-running, steady-state, pin-resolved modeling and simulation tool for analysis of common boiling water reactor (BWR) geometry and common operating conditions. ANTS has been integrated into the Virtual Environment for Reactor Application (VERA) core simulator software, where it can be used to provide a thermal/hydraulic (T/H) subchannel solution that is then used to provide neutronic feedback as well as perform the fuel depletion and temperature solution. This paper presents the rigorous analysis performed on the ANTS code, which includes both code and solution verification testing, benchmarking with the existing two-phase subchannel capability in VERA, CTF, and validation testing using popular two-phase experiments such as PWR Sub-channel and Bundle Tests (PSBT), BWR Full-size Fine-mesh Bundle Tests (BFBT), Risø, and FRIGG. This assessment was used to qualify ANTS for its intended applications before its use for core-scale, multiphysics BWR simulations. In general, it was found that agreement with experimental data was good; errors were within the range of experimental data uncertainty. Furthermore, code and solution verification confirmed that the governing equations and the most important closure terms were correctly implemented and behaving as expected.

A Critical Review of Radiolysis Issues in Water-Cooled Fission and Fusion Reactors: Part I, Assessment of Radiolysis Models

A critical review is presented on modeling of the radiolysis of the coolant water in nuclear power reactors with emphasis on ITER. The review is presented in two parts: In Part I, we assess previous work in terms of compliance with important chemical principles and conclude that no model proposed to date is completely satisfactory, in this regard. Thus, some reactions that have been proposed in various radiolysis models are not elementary in nature and can be decomposed into two or more elementary reactions, some of which are already included in the models. These reactions must be removed in formulating a viable model. Furthermore, elementary reactions between species of like charge are also commonly included, but they can be discounted upon the basis of Coulombic repulsion under the prevailing conditions (T < 350 °C) and must also be removed. Likewise, it is concluded that the current state of knowledge with respect to radiolytic yields (i.e., G-values) is also unsatisfactory. More work is required to ensure that the yields used in radiolysis models are truly “primary” yields corresponding to a time scale of nanoseconds or less. This is necessary to ensure that the impact of the reactions that occur outside of the spurs (ionizing particle tracks in the medium) are not counted twice. In Part II, the authors review the use of the radiolysis models coupled with electrochemical models to predict the water chemistry, corrosion potential, crack growth rate in Type 304 SS, and accumulated damage in the coolant circuits of boiling water reactors, pressurized water reactors, and the test fusion reactor, ITER. Based on experience with fission reactors, the emphasis should be placed on the control of the electrochemical corrosion potential because it is the parameter that best describes the state of corrosion in coolant circuits.

Experimental investigation of liquid entrainment in vertical upward annular flow based on fluorescence imaging

The liquid entrainment in annular flow is essential for the prediction of the critical heat flux (CHF), which is one of the important thermal hydraulics parameters in the nuclear boiling water reactor (BWR). However, few investigations have focused on the liquid entrainment in the annular flow inside pipe. In this paper, the liquid entrainment in the vertical upward gas-liquid annular flow was investigated by the Planar Laser-induced Fluorescence. Five types of entrainment have been observed and studied on the basis of the fluorescence imaging and the quantitative measurements. The frequency of occurrence for each entrainment is investigated, and the influencing factors are further discussed. The results show that for entrainment types 2 and 3, the frequency of occurrence decreases with the increasing of wave frequency, while for the entrainment types 4 and 5 that increases as the fluctuation amplitude of the liquid film increases. Finally, the frequency of occurrence for the five types of entrainment is statistically analyzed, and the influences of flow conditions on the fraction distribution of the entrainment are investigated. With the increase of the gas-liquid superficial velocity, the fraction distribution moves to the left and the right respectively. The transformation of fraction distribution corresponds to the variations in the contributions of entrainment types 1–5 to the total entrainment frequency, which could reveal the essence that the gas-liquid superficial velocity affects the size distributions of liquid entrainments. These results can provide a key solution to investigate the characteristics of liquid entrainment and contribute to better understanding of the liquid entrainment in the gas liquid annular flow.

Development of an Activity Transport Model and in situ Measurements on Deposited Radionuclides in the Primary Coolant Pipelines of the Chinshan Nuclear Power Plant

Activated corrosion products deposited on the surfaces of fuel rods and pipelines contribute the majority of the radiation level in the primary system piping of a light water reactor and would have a significant impact on the safety of maintenance personnel or those involved in future decommissioning work. A computer model for site-specific applications, by the name of ACP_BWR, was developed to predict the distribution of activated corrosion products in the primary coolant circuit of a boiling water reactor (BWR). The prediction results were in reasonably good agreement with the data taken by periodic and in situ measurements at three locations after permanent shutdown of the BWR. Our analyses indicated that the 60Co, 54Mn, 58Co, and 59Fe activities in the core bypass, upper plenum, and lower downcomer regions were higher than those at other regions of the Chinshan Unit 1 reactor. Accordingly, the dose rates resulting from the activated corrosion products deposited at regions close to either side of the core shroud were comparatively high, surpassing those induced by neutron activation at these regions.

Validation and application of numerical modeling for in-vessel melt retention in corium pools

In a hypothetical severe accident scenario in light water reactors (LWR), after the loss of primary coolant, the melted core moves to the lower plenum of the reactor pressure vessel and accumulate there. The melted core cautiously releases the decay heat which forms a pool of melted core, called corium, and turbulent natural convection starts. The decay heat transferred by the corium to the vessel may result in vessel failure due to the focusing effect in the absence of an effective cooling system. A numerical analysis is carried out using STAR-CCM+ commercial software to investigate the ability of the existing turbulence models and Algebraic Heat Flux Model (AHFM) to simulate the natural convective heat transfer phenomena in the corium pools. Efforts have been made to predict the BALI experimental results which are designed to simulate the heat transfer in corium pools, in the framework of severe accident studies. For BALI experiments various test campaigns are run varying the internal Rayleigh number, the viscosity of the simulant fluid, and test facility height. Various such experimental cases are verified and compared with numerical simulations. Temperature profiles along a vertical line, surface heat flux along the curved surface, average Nusselt number on the top, and curved surfaces are estimated. The effect of wall boundary conditions on heat transfer is analyzed. The numerical analysis is extended to Pressurized Water Reactor (PWR) and Boiling Water Reactor (BWR) geometries. The CFD analysis predicts the stratified zone and upper mixing zone very well, and the temperature profiles and average heat transfer values are in agreement with the experiments.

Critical heat flux on zircaloy and accident tolerant fuel cladding under prototypical conditions of pressurized and boiling water reactors

First of a kind high spatial and temporal resolution temperature measurements of Pressurized Water Reactor and Boiling Water Reactor simulated fuel pins (high heat flux cosine profile heaters) in annular rod type geometry during nucleate boiling, departure from nucleate boiling and re-wetting has been performed. Distributed fiber optic measurements of the cladding axial temperature with spatial resolutions down to 2.5 mm axially at up to three equally spaced azimuthal locations per rod have demonstrated the similarities and differences of the different accident tolerant fuel cladding materials. These experiments show the performance of accident tolerant cladding with a precise measurement of the location and evolution of dryout followed by the re-wetting phenomena (see supplementary time history of a critical heat flux event). Experimental parameters varied coolant mass fluxes from 1695 to 2712 kg/m2s, pressures from 10 to 20 MPa, and inlet subcooling from 10 to 55 °C. It was found that the coatings and different material have little effect on the value of the critical heat flux. Any discrepancies between the local critical heat flux verified among the cladding material are attributed to different departure from nucleate boiling phenomena captured during the experiment with the high-resolution optical fiber temperature data. Substantial differences in the performance of the different cladding material to survive a critical heat flux event were however observed by post-test X-ray and optical inspection. External evident damage and internal cracks to the simulated fuel pin having the bare zircaloy cladding were observed in regions associated with the occurrence of the critical heat flux for longer periods. Such damages were associated with zirconium oxidation, and not noticed on the Cr coated zircaloy or FeCrAl claddings. These results indicate that the use of FeCrAl or a Cr coated zircaloy as cladding in Pressurized Water Reactor and Boiling Water Reactor may increase the survivability of the cladding material for short transients.

High temperature electrochemical reaction parameters affecting electrochemical corrosion potential of Ni-base Alloy 182

ABSTRACT High-temperature electrochemical reaction parameters for nickel-base Alloy 182 weld metal (Alloy 182) have been determined based on data measured in pure water at high temperatures to realize an analytical model that can predict electrochemical corrosion potential (ECP) of Alloy 182 without using any adjustment parameters based on ECP data. Behaviors of measured polarization curves for Alloy 182 could be explained well with the determined parameters. Calculated ECPs of Alloy 182 agreed with the measured values in the cases that it was acceptable for the model to have an error range of ±0.1 V in a range from the deaerated condition to 10,000 ppb (=μg kg−1) oxygen concentration. It is expected that the ECP model with the determined electrochemical reaction parameters will have greater validity for, and better interpretability of, the ECPs of Alloy 182 determined in a boiling water reactor environment than previous models.

Brittle fracture initiation in decommissioned boiling water reactor pressure vessel head weld

The brittle fracture initiation behavior and fracture toughness in the ductile-to-brittle transition region for a thermally-aged weld metal of a decommissioned reactor pressure vessel head (in operation at 288 °C for 23 effective full power years) and in the non-aged reference condition were investigated. The results show that brittle fracture initiated primarily from non-metallic inclusions. The correlation between fracture toughness and brittle fracture initiation type (inclusion debonding or breakage), initiator size, initiation location (as-welded or re-heated regions in the weld metal) were analysed. Despite that thermal ageing does not affect significantly the fracture toughness, it could promote the debonding as a brittle fracture primary initiation type. The influence of debonded inclusion on the evolution of cumulative damage and brittle fracture initiation was assessed using crystal plasticity modelling.