Fuel Rod Failure Research Articles

Morphology change of rock-like oxide fuels in reactivity-initiated-accident simulation tests

Pulse irradiation tests under simulated reactivity-initiated accident (RIA) conditions were performed with three types of rock-like oxide (ROX) fuels. Single phase yttria stabilized zirconia (YSZ), homogeneous mixture of YSZ/spinel and YSZ particle dispersed in spinel type ROX fuels were pulse irradiated in the Nuclear Safety Research Reactor (NSRR). Mode and threshold of the fuel rod failure including its consequences were investigated under the RIA conditions. The cladding failure occurred in a burst type mode in all the three types of ROX fuel tests with considerable fuel melting. Even though the mode was quite different from those of UO 2 fuel, failure threshold enthalpies of the ROX fuels were close to that of UO 2 fuel at about 10 GJ m −3. The consequence of the failure of the ROX fuels rods was different from the one of UO 2 fuel rods, because molten fuel dispersal occurred at lower enthalpies in the ROX fuel tests. Change of the fuel structure and material interaction in the transient heating conditions were examined through optical and secondary electron microscopy, and electron probe micro analysis.

Rock-Like Oxide Fuels and Their Burning in LWRs

Research on the plutonium rock-like oxide (ROX) fuels and their once-through burning in light water reactors has been performed to establish an option for utilizing and disposing effectively the excess plutonium. The ROX fuel is a sort of the inert matrix fuels and consists of mineral-like compounds such as yttria stabilized zirconia, spinel and corundum. A particle-dispersed fuel was devised to reduce damage by heavy fission fragments. Some preliminary results on swelling, fractional gas release and microstructure change for five ROX fuels were obtained from the irradiation test and successive post-irradiation examinations. Inherent disadvantages of the Pu-ROX fuel cores could be improved by adding 238U or 232Th as resonant materials, and all improved cores showed a nearly the same characteristics as the conventional UO2 core during transient conditions. The threshold enthalpy of the ROX fuel rod failure was found to be comparable to the fresh UO2 rod by pulse-irradiation tests simulating reactivity initiated accident conditions.

A Statistical Approach to Predict the Failure Enthalpy and Reliability of Irradiated PWR Fuel Rods During Reactivity-Initiated Accidents

During the last decade, the failure behavior of high-burnup fuel rods under a reactivity-initiated accident (RIA) condition has been a serious concern since fuel rod failures at low enthalpy have been observed. This has resulted in the reassessment of existing licensing criteria and failure-mode study. To address the issue, a statistics-based methodology is suggested to predict failure probability of irradiated fuel rods under an RIA. Based on RIA simulation results in the literature, a failure enthalpy correlation for an irradiated fuel rod is constructed as a function of oxide thickness, fuel burnup, and pulse width. Using the failure enthalpy correlation, a new concept of “equivalent enthalpy” is introduced to reflect the effects of the three primary factors as well as peak fuel enthalpy into a single damage parameter. Moreover, the failure distribution function with equivalent enthalpy is derived, applying a two-parameter Weibull statistical model. Finally, the sensitivity analysis is carried out to estimate the effects of burnup, corrosion, peak fuel enthalpy, pulse width, and cladding materials used.

The modeling of fuel rod behaviour under RIA conditions in the code DYN3D

A description of the fuel rod behaviour and heat transfer model used in the code DYN3D for nuclear reactor core dynamic simulations is given. Besides the solution of heat conduction equations in fuel and cladding, the model comprises a detailed description of heat transfer in the gas gap by conduction, radiation and fuel-cladding contact. The gas gap behaviour is modeled in a mechanistic way taking into account transient changes of the gas gap parameters based on given conditions for the initial state. Thermal, elastic and plastic deformations of fuel and cladding are taken into account within 1D approximation. A creeping law for time-dependent estimation of plastic deformations is implemented. Metal–water reaction of the cladding material in the high temperature region is considered. The cladding–coolant heat transfer regime map covers the region from one-phase liquid convection to dispersed flow with superheated steam. Special emphasis is put on taking into account the impact of thermodynamic non-equlibrium conditions on heat transfer. For the validation of the model, experiments on fuel rod behaviour during RIAs carried out in Russian and Japanese pulsed research reactors with shortened probes of fresh fuel rods are calculated. Comparisons between calculated and measured results are shown and discussed. It is shown, that the fuel rod behaviour is significantly influenced by plastic deformation of the cladding, post crisis heat transfer with sub-cooled liquid conditions and heat release from the metal–water reaction. Numerical studies concerning the fuel rod behaviour under RIA conditions in power reactors are reported on. It is demonstrated, that the fuel rod behaviour at high pressures and flow rates in power reactors is different from the behaviour under atmospheric pressure and stagnant flow conditions in the experiments. The mechanisms of fuel rod failure for fresh and burned fuel reported from the literature can be qualitatively reproduced by the DYN3D model. However, the model must be extended and improved for a proper description of burned fuel behaviour. The realistic simulation of the fuel rod behaviour is important not only under RIA conditions, but also for the analysis of operational transients. This has been shown in a calculation of an operational transient with power decrease after switching-off one from the two working feed water pumps in the NPP Balakovo (VVER-1000).

Deformation–corrosion interactions for Zr alloys during I-SCC crack initiation: Part II: Localised stress and strain contributions

For a better understanding of the initiation step of iodine induced stress corrosion cracking (SCC) in Zr alloys, responsible for pellet–cladding interaction (PCI) fuel rod failures, an analytical study has been undertaken, the aim of which being focused on the respective roles of local chemistry and stress/strain state on the crack nucleation. This second part is mostly related to the local stress induced by strain incompatibilities between grains. Using EBSP (electron back-scattering pattern) to analyze the crystallographic orientation of all the grains of the samples tested in SCC, it was possible to conclude that the major parameter controlling the nucleation of the intergranular cracks is not related to grain to grain strain incompatibilities, but to the orientation of the grain boundary planes with respect to the tensile stress.

Deformation–corrosion interactions for Zr alloys during I-SCC crack initiation: Part I: Chemical contributions

For a better understanding of the initiation step of iodine-induced stress corrosion cracking in Zr alloys, responsible for pellet–cladding interaction (PCI) fuel rod failures, an analytical study has been undertaken, the aim of which being focused on the respective roles of local chemistry and stress/strain state on the crack nucleation. This first part is mostly related to the chemical environment. From the tensile tests performed under iodine rich and inert environments, it was concluded that no crack initiation could be detected following the tests in an inert atmosphere. The iodine induced stress corrosion initiation mechanism must therefore be analysed as a corrosion–strain interaction.

98/02951 Development of a computer code to estimate the fuel rod failure using primary coolant activities of operating PWRs

98/02951 Development of a computer code to estimate the fuel rod failure using primary coolant activities of operating PWRs

Development of a computer code to estimate the fuel rod failure using primary coolant activities of operating PWRS

A Windows computer code entitled as “CAAP (Cooland Activity Analysis Program)” has been developed to evaluate the number, the degree of failures, and the location of failed fuel rods using primary coolant radioactivity data obtained from operating PWRS. New models are developed and incorporated into the CAAP program to improve some of the drawbacks of existing computer codes. The iodine and noble gas activities obtained by grab sampling and that obtained from the primary coolant radioactivity monitoring system can be used to estimate fuel rod failures with CAAP. For an on-line evaluation of the number of failed fuel rods and the degree of failures, CAAP has been combined with the primary coolant radioactivity monitoring system at Kori 3 & 4 operating nuclear power plants. The validity of the computational models of CAAP has been examined using nuclear power plant data collected. For 29 cycles of PWRs for which ultrasonic inspections were performed at the end of fuel cycles, the number of failed fuel rods has been estimated with CAAP and compared with the ultrasonic inspection results. The results show that CAAP gives a better agreement with the ultrasonic test data than existing models. In addition, CAAP calculations performed to estimate the region and the burnup of the failed fuel rods for 2 cycles, whose locations of failed fuel rods were known, show that CAAP can predict both the region and the burnup of the failed fuel rods accurately.

Fuel failure and fission gas release in high burnup PWR fuels under RIA conditions

To study the fuel behavior and to evaluate the fuel enthalpy threshold of fuel rod failure under reactivity initiated accident (RIA) conditions, a series of experiments using pulse irradiation capability of the Nuclear Safety Research Reactor (NSRR) has been performed. During the experiments with 50 MWd/kg U PWR fuel rods (HBO test series; an acronym for high burnup fuels irradiated in Ohi unit 1 reactor), significant cladding failure occurred. The energy deposition level at the instant of the fuel failure in the test is 60 cal/g fuel, and is considerably lower than those expected and pre-evaluated. The result suggests that mechanical interaction between the fuel pellets and the cladding tube with decreased integrity due to hydrogen embrittlement causes fuel failure at the low energy deposition level. After the pulse irradiation, the fuel pellets were found as fragmented debris in the coolant water, and most of these were finely fragmented. This paper describes several key observations in the NSRR experiments, which include cladding failure at the lower enthalpy level, possible post-failure events and large fission gas release.

A model for predicting pellet-cladding interaction-induced fuel rod failure

A model for predicting pellet-cladding mechanical-interaction-induced fuel rod failure is presented. Cladding failure is predicted by explicitly modelling the formation and propagation of radial cladding cracks by the use of non-linear fracture mechanics concepts in a finite element computational framework. The failure model is intended for implementation in finite element fuel performance codes in which local pellet-clad interaction is modelled. Crack initiation is supposed to take place at pre-existing cladding flaws, the size of which is estimated by simple probabilistic concepts, and the subsequent crack propagation is assumed to be due to either iodine-induced stress corrosion cracking or ductile fracture. The novelty of the outlined approach is that the development of cladding cracks which may ultimately lead to fuel rod failure can be treated as a dynamic and time-dependent process. The influence of complex or cyclic loading, ramp rates and material creep on the failure mechanism can thereby be investigated. The presented failure model has been incorporated in the ABB Atom stav-t transient fuel performance code. Numerical results from some applications of the code are used to illustrate the usefulness of the model.

Radiochemical Evaluation for Debris-Induced Failures

Radiochemical trends and anomalies experienced during cycle 15 of the Haddam Neck nuclear power plant, as a result of > 450 debris-induced fuel rod failures, presented a situation previously unreported in the nuclear industry. These data, along with shutdown and depressurization spiking data, needed to be evaluated against ultrasonic fuel assembly examination results to derive a predictive model, called the xenon pin equivalent (XPE), to be used for cycle 16. During the development of the model, a fission product release mechanism for this particular type of failure needed to be postulated based on cycle 15 data. The predictive model was tested during cycle 16, which presented similar but more subtle radiochemical trends than cycle 15. Several operational events affected the XPE model, including use of degasification and down-power maneuvers. After the cycle 16 shutdown, the XPE model results were reviewed and evaluated against ultrasonic testing results. Although expected to be conservative, this evaluation proved encouraging in that the model performed more accurately than expected. Additionally, these data helped confirm the postulated release mechanism and its contribution to the XPE model.

Development of a solution to the debris fretting problem

One of the major recognized causes of fuel rod failures is fretting of the clad due to the entrapment of small particles of debris in a fuel rod spacer. Such debris, inadvertently dropped into the primary system during maintenance operations, includes various sizes of particles. Very small particles of the size of a grain of sand pass through the system without harm. Some larger size particles such as large bolts and nuts cannot pass through the fuel assembly inlet (lower tie plate) and thus cause no harm to the fuel rods. Intermediate size particles, however, such as metal cuttings, electrical connectors, metal fittings, pieces of wire, and small nuts and bolts can pass through the holes in the lower tie plate and become trapped between fuel rods in a spacer where hydrautically induced vibrations can cause fretting failure of the fuel rod. A Curved-Blade Tie Plate (FUELGARD™) has been developed which effectively traps the intermediate size debris before it enters the lower spacer. This specially designed tie plate greatly reduces the danger of debris damage with little or no increase in the pressure drop of the coolant flow system.

A comprehensive in-pile test of PWR fuel bundle

An in-pile test of PWR fuel bundle has been conducted in HWRR at IAE of China. This paper describes the structure of the test bundle (3 × 3−2), fabrication process and quality control of the fuel rod, irradiation conditions and the main Post Irradiation Examination (PIE) results. The test fuel bundle was irradiated under the PWR operation and water chemistry conditions with an average linear power of 381 W/cm and reached an average burnup of 25010 MWd/tU of the fuel bundle. After the test, destructive and non-destructive examination of the fuel rods was conducted at hot laboratories. The fission gas release was 10.4–23%. The ridge height of cladding was 3 to 8 μm. The hydrogen content of the cladding was 80 to 140 ppm. The fuel stack height was increased by 2.9 to 3.3 mm. The relative irradiation growth was about 0.11 to 0.17% of the fuel rod length. During the irradiation test, no fuel rod failure or other abnormal phenomena had been found by the on-line fuel failure monitoring system of the test loop and water sampling analysis. The structure of the test fuel assembly was left undamaged without twist and detectable deformation.

In-core surveillance of defective PWR fuel elements in the case of fuel-to-water contact

A large-scale experiment on defective fuel rods was carried out in a PWR. Non-destructive and destructive post irradiation examinations on one of the test assemblies with the artificial fuel cladding defects revealed secondary fuel rod failures which resulted in fuel release to the primary circuit. It is shown that the presence of fuel rod failures with fuel-to-water contact induces a special situation for fuel rod integrity surveillance during reactor operation. Higher uranium contamination levels on the in-core surfaces of the fuel rods presented complications for the assessment of their condition; this aspect, together with the means used to recognize the true state of the rods — despite these complications — are discussed.

Validation of frey for the safety analysis of LWR fuel using transient fuel rod experiments

The analysis and comparison of severe light water reactor transient experiments are presented from the FREY verification and validation effort. The purpose of this study was to validate the predictive capabilities of the code for severe transient analysis. The FREY code, developed under the sponsorship of the Electric Power Research Institute, uses a two-dimensional finite-element computational method for the thermomechanical analysis of LWR fuel rods under steady state and transient conditions. A total of 10 test fuel rods from experimental programs conducted in both the Power Burst Facility and the Transient Reactor Test Facility have been used in this study. The fuel rods were selected from the following test programs: Power Coolant Mismatch Tests, PCM-2 and PCM-4: Reactivity Initiated Accident Test, RIA 1–2; Loss-of-Coolant Accident Test, LOC-3; First Fuel Rod Failure Test, FRF-1; and Irradiation Effects Test, IE-3. The test programs used in this study cover a large range of code applications for severe transient analysis. The methods used to model the fuel, cladding, and coolant geometry are discussed in addition to experimental data comparisons. The results of the PCM-2, RIA 1–2, and FRF-1 analyses are presented to highlight the full two-dimensional modeling capabilities of FREY and to compare the thermal and mechanical measurements with FREY's prediction. The comparisons show good general agreement, with a tendency for FREY to overpredict the peak cladding surface temperature for a few cases where strong three-dimensional effects have been identified.

Experiments on the load following behaviour of PWR fuel rods

KWU had studied the effects of load following operation on fuel performance from the beginning of commercial operation of nuclear power plants: The first power cycling experiments were started in 1970 in the nuclear power plant Obrigheim (KWO) and in the High Flux Reactor (HFR) Petten. These power cycling tests performed at various power levels and burnups of up to 25 GWd/t(U) showed that the fuel rod cycling performance compares well with the performance of fuel rods operated under essentially constant load at comparable power levels. Two additional cycling tests as described in this paper were performed in the HFR Petten with preirradiated PWR fuel rods having burnups of up to 40 GWd/t(U). These experiments comprised up to 60 cycles between 250/360 W/cm and 215/320 W/cm with 10% power overshoot (400, 370 W/cm) after each cycle. Also, these experiments ended up with sound fuel rods. Moreover, detailed investigations before and between power cycles and after experiment termination showed clearly that the fuel performance corresponds to a single ramp to peak power and that the cycling effects are indeed very small. This confirmed earlier findings that due to crack reversal in the UO 2 the cyclic dimensional changes mainly occur in the UO 2 itself. Altogether the experiments show that power cycling does not lead to fuel rod failures, which is also confirmed by successful load follow operation in commercial power plants.

Studies of the UO 2-zircaloy chemical interaction and fuel rod relocation modes in a severe fuel damage accident

Experiments have been conducted in the Nuclear Safety Research Reactor (NSRR) at JAERI since 1975 in order to study fuel rod failure behavior under reactivity-initiated accident conditions. Recently the experiments have been focussed on fuel behavior under simulated severe fuel damage (SFD) accident conditions. UO 2-Zircaloy reaction kinetics during very rapid transients at elevated temperatures was studied from a metallurgical point of view. Equilibrium was found to be established even in very rapid transients. The reaction rate equations developed in isothermal studies can be applied to interpret the experimental results. A fuel rod relocation criterion in connection with peak temperatures, environment conditions and initial fuel rod conditions was developed. According to the test results, fuel rod melt down due to liquefaction seems unlikely below the melting temperature of β-Zircaloy.

Cladding Embrittlement and Fuel Rod Failure Threshold under Reactivity Initiated Accident Condition

Fuel rod failure behavior has been studied under a reactivity initiated accident condition in Nuclear Safety Research Reactor (NSRR), JAERI. In the studies, inetallurgical observations showed that the incipient fuel rod failure mode was oxygen-induced embrittlement of the cladding independent of the test conditions such as fuel designs and cooling environments except for pressurized and waterlogged fuels. Development of the oxidation layers and embrittlement of β-Zry were quantitatively evaluated through the metallurgical examinations. A diffusion equation of oxygen was solved under a finite system with moving boundary conditions to obtain the oxygen concentration and evaluate the cladding embrittlement. The calculation showed that the wall thinning due to the cladding melt is needed for the complete embrittlement because the wall thinning enhances the oxygen concentration in the β-Zry, which well explain the experimental results. Therefore the failure threshold energy is determined by the cladding melting temperature. The failure threshold derived from this study is expected to be applicable to predicting the fuel rod failure behavior in computer analyses and also useful to evaluate the failure threshold energy for the new types of fuel rod.

Cladding embrittlement and fuel rod failure threshold under reactivity initiated accident condition.

Fuel rod failure behavior has been studied under a reactivity initiated accident condition in Nuclear Safety Research Reactor (NSRR), JAERI. In the studies, inetallurgical observations showed that the incipient fuel rod failure mode was oxygen-induced embrittlement of the cladding independent of the test conditions such as fuel designs and cooling environments except for pressurized and waterlogged fuels. Development of the oxidation layers and embrittlement of β-Zry were quantitatively evaluated through the metallurgical examinations. A diffusion equation of oxygen was solved under a finite system with moving boundary conditions to obtain the oxygen concentration and evaluate the cladding embrittlement. The calculation showed that the wall thinning due to the cladding melt is needed for the complete embrittlement because the wall thinning enhances the oxygen concentration in the β-Zry, which well explain the experimental results. Therefore the failure threshold energy is determined by the cladding melting temperature. The failure threshold derived from this study is expected to be applicable to predicting the fuel rod failure behavior in computer analyses and also useful to evaluate the failure threshold energy for the new types of fuel rod.

Failure Behavior of Stainless Steel Clad Fuel Rod under Simulated Reactivity Initiated Accident Condition

In-reactor experiments were performed in Nuclear Safety Research Reactor of Japan Atomic Energy Research Institute to study the failure behavior of stainless steel clad fuel rods under a simulated reactivity initiated accident (RIA) condition. A single test fuel rod with stainless steel cladding was contained in a capsule filled with water at room temperature and atmospheric pressure and irradiated by pulsing power simulating an RIA. It was revealed through the experiments that the failure mechanism of the stainless steel clad fuel rod was cladding melting, which was different from oxygen-induced embrittlement observed in the Zircaloy clad fuel rod in the same test condition, and the failure threshold energy was determined to be about 240cal/g·UO2 (–1,000 kJ/kg·UO2), which was about 20 cal/g·UO2 (–85 kJ/kg·UO2) lower than that of the Zircaloy clad fuel rod. It was also found that the mechanical energy was generated by explosive vaporization of coolant due to molten fuel-coolant interaction as a consequence of the fuel rod failure accompanying fuel pellet fragmentation at an energy deposition of nearly 380 cal/g·UO2 (–1,600 kJ/kg·UO2) or more.