A creep rupture criterion for Zircaloy-4 fuel cladding under internal pressure

The usual criterion which limits the cladding strain to 0.01 to prevent the creep rupture under internal pressure seems too conservative for application to transport and interim storage. So we have analysed CEA’s data on this subject for CWSR Zircaloy-4 in order to find a less conservative criterion. Temperatures between 350 and 470 °C were studied for stresses between 100 and 550 MPa, according to the irradiation level from 0 to 9.5 × 10²⁵ n m⁻². Except for high stressed irradiated material (because of low ductility), the plastic instability appears as the major mechanism of rupture. For the unirradiated material, it is essentially due to the stress increase with strain. This instability is accelerated by annealing for the irradiated one at moderate or low stress. From these considerations, we propose a new rupture criterion for CWSR Zircaloy-4 cladding submitted to internal pressure, for both unirradiated and irradiated materials.     

Zircaloy-4 cladding corrosion model covering a wide range of PWR experiences

A phenomenological corrosion model for Zircaloy-4 cladding was developed by focusing on the effect of the metallurgy of cladding and the water chemistry combined with the thermo-hydraulic conditions. The metallurgical effect was formulated by considering the Sn content in the cladding and the heat treatment of the cladding. Concerning the effect of the water chemistry, it is assumed that lithium and boron have an influence on the corrosion under the condition of subcooled void formation on the cladding surface. The developed corrosion model was implemented in a fuel performance code, COSMOS, and verified using the database obtained for the UO₂ and MOX fuel rods irradiated in various PWRs. It was elucidated that the corrosion by lithium was enhanced in the case where the fuel rods were irradiated with a high linear power so that a significant subcooled void could be formed on the cladding surface. On the other hand, there was no evidence of the lithium effect even though its concentration was high enough if the void in the coolant was negligible. This result shows that the acceleration of corrosion by an increased lithium concentration occurs only when subcooled voids are formed on the cladding surface. In addition, the comparison between the measurement and the prediction for the MOX fuel rods indicates that no distinguishable difference is found in the corrosion behavior between the MOX and the UO₂ fuels as expected.     

Effect of Hydrogen Water Chemistry on Zircaloy-2 Fuel Cladding and Structure Material Performance

The effects of hydrogen addition to the feedwater on the corrosion and hydrogen uptake performance of Zircaloy-2 fuel cladding tubes, a water rod tube and spacer materials irradiated for four cycles in a BWR were evaluated. The uniform oxide behaviors of the cladding tubes, water rod and spacer materials were not affected by hydrogen water chemistry (HWC) condition. The hydrogen uptake and pickup fractions of the water rod and spacer materials were similar to those of water rods and spacer materials under normal water chemistry (NWC) conditions. As for the fuel rods, in spite of comparably heavy crud deposition, their hydrogen uptake and pickup fractions were clearly lower than the values under NWC conditions. Overall, the results indicated that HWC had no adverse effects on fuel performance.     

Behavior of 60 to 78MWd/kgU PWR Fuels under Reactivity-Initiated Accident Conditions

To provide a data base for the regulatory guide of light water reactors, behavior of reactor fuels during off-normal and postulated accident conditions such as reactivity-initiated accident (RIA) is being studied in the Nuclear Safety Research Reactor (NSRR) program of the Japan Atomic Energy Agency (JAEA). The paper presents recent results obtained from the NSRR power burst experiments with high burnup fuels, and discusses effects of pellet expansion as PCMI (Pellet-Cladding Mechanical Interaction) loading and cladding embrittlement primarily due to hydrogen absorption. Results from the recent four experiments on high burnup (about 60 to 78 MWd/kgU) PWR UO₂ rods with advanced cladding alloys showed that the fuel rods with improved corrosion resistance have larger safety margin against the PCMI failure than conventional Zircaloy-4 rods. The tests also suggested that the smaller inventory of inter-granular gas in the pellets with the large grain could reduce the fission gas release during the RIA transient; and high burnup structure in pellet periphery (so-called rim structure) does not have strong effect on reduction of the failure threshold because the PCMI load is produced primarily by solid thermal expansion.     

Embrittlement behavior of zircaloy-4 cladding during oxidation and water quench

Simulated loss of coolant accident (LOCA) tests on Zircaloy-4 cladding were carried out to evaluate the thermal shock property during the injection of emergency core coolant. A Zircaloy-4 specimen was oxidized in a steam environment between 1000 and 1250 °C followed by a flooding of the cooling water. After the test, the ductility of the thermally embrittled specimen was measured by a ring-compression test and a microstructural analysis was carried out. The results showed that the threshold equivalent cladding reacted (ECR) value to cause a failure was higher than the conventional 17% criterion calculated by the Baker–Just equation. A residual metal thickness under 0.3 mm as well as a ring-compression ductility below 0.2 mm for a fracture is effective to assess the thermal shock embrittlement of Zircaloy-4 in an axially unrestrained, or even in a restrained condition.     

Failure-Bearing Capability of Oxidized Zircaloy-4 Cladding under Simulated Loss-of-Coolant Condition

With a view to examining the failure-bearing capability of Zircaloy-4 cladding under postulated Loss-of-Coolant Accident condition in LWRs, integral tests of rod-burst, oxidation and thermal-shock were performed using simulated fuel containing A1₂0₃ pellets sheathed in Zircaloy-4 specimen cladding, filled with He gas, and sealed. This simulated fuel rod was oxidized in steam flowing at the isothermal oxidation temperatures between 920 and 1,330°C for duration ranging of 3~180 min after the cladding burst. After isothermal oxidation, the rod was quenched with bottom-flooding water under the condition of constraint or no constraint.

The failure boundary oxidation condition of the cladding on quenching under no constraint condition lay in the region of 35~38% ECR for the isothermal oxidation temperatures between 1,050 and 1,330°C. For the temperatures ranging 970~1,050°C, the boundary value of ECR was somewhat lower than that obtained for higher temperatures.

The failure boundary oxidation condition of the cladding on quenching under constraint condition lay in the region of 19~24% ECR for the isothermal oxidation temperatures between 930 and 1,310°C. It is sufficiently large compared with the criterion of 15% ECR in Japanese acceptance criteria for ECCS. Hydrogen absorbed by the Zircaloy-4 cladding as well as oxygen played a dominant role in the fracture behavior of the rod during flooding under constraint condition.     

The anisotropy of the high temperature deformation of Zircaloy-cladding tubes for nuclear fuel rods

The anisotropy of the high temperature deformation of Zircaloy-4 cladding tubes for nuclear fuel rods for pressurized water reactors has been investigated. The axial and tangential components of the deformation of internally pressurized tube samples during closed end creep rupture tests in air at 800°C have been measured. An axial contraction of the tube sample is observed. Using Hill's theory of plasticity the axial strain can be described by anisotropy coefficients which depend on the texture of the tube material. The anisotropy coefficients are quantitatively related to the orientation distribution of the basal poles in the radial/tangential plane of the tube sample. For the typical texture of Zircaloy cladding tubes of nuclear fuel rods for pressurized water reactors, an axial contraction has to be expected under the biaxial stress conditions applied.     

Experimental and calculation results of the integral reflood test QUENCH-15 with ZIRLO™ cladding tubes in comparison with results of previous QUENCH tests

The QUENCH-15 experiment investigated the effect of ZIRLO™¹ cladding material on bundle oxidation and core reflood, in comparison with tests QUENCH-06 (standard Zircaloy-4), QUENCH-12 (VVER, E110), and QUENCH-14 (M5^®). The QUENCH-15 bundle cross-section corresponded to a Westinghouse PWR core design and consisted of 24 heated rods (internal tungsten heaters between 0 and 1024 mm axial elevation, cladding oxidised region between −470 and 1500 mm), six corner rods made of Zircaloy-4, two corner rods made of E110, and a Zirconium 702 shroud. The test was conducted in principle with the same protocol as QUENCH-06, -12 and -14, so that the effects of the change of cladding material and bundle geometry could be more easily observed. The test protocol involved pre-oxidation to a maximum of about 150 μm oxide thickness at a temperature of about 1473 K over a period of about 3000s. The power was then ramped at a rate of 0.25 W/s/rod to cause a temperature increase until the desired maximum bundle temperature of 2153 K was reached. The maximum oxide layer thickness observed was 380 μm. Then reflood with 1.3 g/s/rod water at room temperature was initiated. The electrical power was reduced to 175 W/rod during the reflood phase, approximating effective decay heat level. The post-test metallography of the bundle showed neither noticeable breakaway oxidation of the cladding nor melt release into space between rods. The average outer oxide layer thickness at hottest elevation of 950 mm was 620 μm (QUENCH-06: 630 μm). The molten cladding metal at hottest elevation was localised between the outer and inner oxide layers. The thickness of inner oxide layer reaches 20% of that of the outer oxide layer. The measured hydrogen release during the QUENCH-15 test was 41 g in the pre-oxidation and transient phases and 7 g in the quench phase which are comparable with those in QUENCH-06, i.e. 32 g and 4 g, respectively. Post-test calculations were performed using a version of SCDAP/RELAP5/MOD3.2. The calculation results support the heuristic observation that there was no major difference between the influence of Zircaloy-4, M5^® or ZIRLO™ for the beyond-design basis accident present conditions here studied.     

Experimental and calculation results of the integral reflood test QUENCH-14 with M5® cladding tubes

The QUENCH-14 experiment investigated the effect of M5® cladding material on bundle oxidation and core reflood, in comparison with tests QUENCH-06 (ISP-45) that used standard Zircaloy-4 and QUENCH-12 that used VVER E110-claddings. The PWR bundle configuration of QUENCH-14 with a single unheated rod, 20 heated rods, and four corner rods was otherwise identical to QUENCH-06. The test was conducted in principle with the same protocol as QUENCH-06, so that the effects of the change of cladding material could be observed more easily. Pre-test calculations were performed by the Paul Scherrer Institut (Switzerland) using the SCDAPSIM, SCDAP/RELAP5 and MELCOR codes. Follow-on post-test analyses were performed using SCDAP/RELAP5 and MELCOR as part of an ongoing programme of model validation and code assessment. Alternative oxidation correlations were used to examine the possible influence of the M5® cladding material on hydrogen generation, in comparison with Zircaloy-4.     

Failure Propensities of Pressurized Zircaloy Tube Containing Iodine and Other Chemical Species

Abstract

Failure propensities of Zircaloy-4 cladding tube internally pressurized with Ar gas containing iodine and iodine plus each of other chemical species were examined at 360°C, to study the effect of corrosive fission products (FPs) on the integrity of spent nuclear fuel rods during dry storage, and also to assess the capability of preventing the spent fuel rod degradation.

The iodine stress corrosion cracking (SCC) of Zircaloy tube occurred in the long time/low stress exposure tests at stresses much lower than the conventional “threshold stress”, with considerably large strains at failure. The addition of cesium to iodine perfectly suppressed the SCC. It is inferred from these results that the degradation of spent fuel rods induced by corrosive fission products is unlikely during dry storage. Even if iodine alone should take effect, a proper strain limit could prevent spent fuel rods from incurring iodine induced effects because of considerably large strains necessary for iodine SCC of Zircaloy tube at low stresses.     

Changes in chemical structure of oxidation reaction layers of Zircaloy-4 and Ti by micro X-ray diffractometry

A micro X-ray diffractometer with a micrometer sized beam concentrator was developed to investigate the changes in the chemical structures of oxide layers for Zr-based alloys (Zircaloy-4) and Ti metal from the center of the cross section to the surface. Zircaloy-4 and Ti metal were chosen because of their use as a fuel cladding and a heat exchange tubing in a nuclear reactor, respectively. The diffraction patterns were obtained from the cross sectional specimens of the oxidized Zircaloy-4 and Ti metal at 50 μm intervals. For the cross section of Zircaloy-4, Zr metal (hexagonal) was identified in the center, ZrO_2−x (hexagonal, about 200 μm in thickness) inside the edge and ZrO₂ (monoclinic, about 400 μm in thickness) at the edge. In the case of Ti metal, Ti metal (hexagonal) was identified in the center, TiO (cubic, about 200 μm in thickness) inside the edge and rutile-TiO₂ (tetragonal, about 230 μm in thickness) at the edge. From this study, it was concluded that the intermediate phase formed between the fuel and the cladding can be identified by the micro-XRD system.     

Effect of Absorbed Hydrogen on the Stress Corrosion Cracking (SCC) Susceptibility of Unirradiated Zircaloy Cladding

Effect of absorbed hydrogen on the stress corrosion cracking (SCC) susceptibility of unirradiated Zircaloy cladding was examined. The data obtained from literatures showed that the normalized ratios of SCC threshold stress (σ_th ) to 0.2% yield stress (σ_0.2.) claddings, from which the influence of σ_0.2 had been eliminated, increased with increasing hydrogen contents below 50ppm in unirradiated Zircaloy-2 and -4. For Zircaloy-4, the break point was observed in the relationship between normalized ratios of σ_th to σ_0.2 and hydrogen content in sample at hydrogen content of approximately 50ppm. Thermodynamic calculations were carried out for the reaction between iodine gas and zirconium containing hydrogen. The results suggested that the reactions hardly occurred at increased hydrogen content and zirconium reacted with iodine gas only below 100 ppm of hydrogen. Since these tendencies correspond to those of the normalized ratios of σ _th to σ_0.2 on the hydrogen content, it is considered that hydrogen affects the reactions between iodine gas and zirconium and reduces the SCC susceptibility of Zircaloy cladding.     

In-Pile Eddy Current Test on PWR Fuel Rod Failed by Pellet-Cladding Interaction

Zirconium oxide nodules formed on BWR fuel rods were characterized quantitatively and correlated statistically with the rod operational parameters. Cladding specimens were obtained from fuel rods irradiated in a commercial BWR. Their burnup and fast neutron fluence ranged 17~38 GWd/t and 4×10²⁵~8×10²⁵ n/m², respectively. Characteristic variables of the nodules such as maximum thickness T ^max (μm) were measured on metallographs of the cladding cross sections. These variables were correlated by multiple regression analyses with the operational parameters, such as irradiation time t (d), linear heat rate p (kW/m) and fast neutron flux ø (n/m²-s). For example, the maximum thickness depended on linear heat rate and showed a saturating tendency with burnup B (GWd/t) (T^max ∝ t0.8+0.5 p2.3±0.9 or T ^max ∝B0.8+0.4p1.5±0.5). This decrease of growth rate with irradiation time was interpreted in terms of a microstructure change of Zircaloy-2 during neutron irradiation. Results of transmission microscopy and energy dispersive X-ray spectroscopy indicated that the alloying elements such as Fe, Cr and Ni dissolved from intermetallic precipitates into the base metal during neutron irradiation. Dissolution of the alloying elements might be effective in decreasing the growth rate of nodules.     

Behavior of High Burn-up Fuel Cladding under LOCA Conditions

LOCA-simulated experiments were performed with MDA, ZIRLO?, M5®, NDA, and Zircaloy-2 cladding specimens with local burn-ups ranging from 66 to 76 MWd/kg. Short test rods fabricated with the cladding specimens were heated, isothermally oxidized at 1,459 to 1,480K in steam flow, and finally quenched in flooding water. Rod rupture and subsequent double-sided oxidation of the cladding were also simulated in the experiments. Neither split-fracture nor fragmentation occurred during the quench in the cladding specimens which were oxidized to about 18–27% of the metallic thickness. Accordingly, the fracture boundary, a most important safety issue, is not reduced significantly by the high burn-up and use of the new alloys within the examined scope, although it may be somewhat reduced with pre-hydriding during the reactor operation as observed in unirradiated specimens.     

Zircaloy-4 Cladding Embrittlement due to Inner Surface Oxidation under Simulated Loss-of-Coolant Condition

Embrittlement of Zircaloy-4 cladding by oxidation of the inner surface occurring in an LWR loss-of-coolant accident was studied using simulated fuel containing of A1₂O₃ pellets sheathed in Zircaloy-4 specimen cladding, filled with Ar gas, and sealed. This simulated fuel rod was heated from outside until the isothermal oxidation temperature between 880 and 1,167°C was obtained after the cladding burst. This exposed the inner surface of the cladding to the environmental atmosphere, provided by steam flowing at a constant rate in the range of 0.13–1.6 g/cm²-min.

The embrittlement of the specimen due to inner surface oxidation is influenced primarily by the amount of hydrogen absorbed by the Zircaloy-4. Ring compression tests conducted at 100°C on test pieces constituted of sliced sections of oxidized specimen showed that Zircaloy containing more than 200–300 wt.ppm of absorbed hydrogen became brittle when oxidized at temperatures above 1,000°C. In the range of oxidation temperature 932 to 972°C, brittleness did not appear below 500–750 wt.ppm absorbed hydrogen.

Hydrogen absorbed by the Zircaloy precipitated in the form of fine hydride crystals formed along previous β-phase grain boundaries. Peaks were found in the distribution of hydrogen absorbed on the inner surface, at a distance of 15–45 mm upward and downward of the rupture opening. Within this range, the distance was influenced by the oxidation temperature and steam flow rate.     

Development of the Ultra-microhardness Technique for Post Irradiation Examination of Fuel Cladding Tubes

Hardness measurements are potentially valuable for a quantitative discussion of embrittlement in the inner portions of fuel cladding tubes. The size of the indentation, however, is not negligible compared to the measuring region, even when a micro Vickers hardness tester is employed. This limits the measuring technique, and very little has been studied about degradation phenomena in the inner portion of the tubes.

A hardness measurement system, equipped with a depth-sensing indentation instrument, and the necessary post irradiation examination technique for specimens with high radioactivity were successfully developed and the following observations were obtained from the system's application example. The diffusion coefficient of oxygen obtained from the hardness of an unirradiated zirconium lined cladding with simulated oxidation in the fuel rod showed good agreement with literature data. The calculated diffusion coefficient from hardness in the inner portion of irradiated Zircaloy-2 fuel rods was almost the same value as that of unirradiated zirconium, which implied that neither neutron irradiation nor fission fragment bombardment enhanced the oxygen diffusion in the inner portion of cladding tube.     

Fracture limit of high-burnup advanced fuel cladding tubes under loss-of-coolant accident conditions

ABSTRACT

To evaluate the fracture limit of high-burnup advanced fuel cladding tubes under loss-of-coolant accident (LOCA) conditions, laboratory-scale integral thermal shock tests were performed using the following advanced fuel cladding tubes with burnups of 73–85 GWd/t: M-MDA^TM, low-tin ZIRLO^TM, M5®, and Zircaloy-2 (LK3). In total eight integral thermal shock tests were performed for these fuel cladding tube specimens, simulating LOCA conditions including ballooning and rupture, oxidation, hydriding, and quenching. During the tests, the specimens were oxidized to 10%–30% equivalent cladding reacted (ECR) at approximately 1473 K and were quenched under axial restraint load of approximately 520–530 N. The effects of burnup extension and use of the advanced fuel cladding tubes on the ballooning and rupture, oxidation, and hydriding under LOCA conditions were inconsiderable. Further, the high-burnup advanced fuel cladding tube specimens did not fracture in the ECR values equal to or lower than the fracture limits of the unirradiated Zircaloy-4 cladding tube reported in previous studies. Therefore, it can be concluded that the fracture limit of fuel cladding tubes is not significantly reduced by extending the burnup to approximately 85 GWd/t and using the advanced fuel cladding tubes, though it slightly decreases with increasing initial hydrogen concentration.     

Stress corrosion cracking of Zircaloy-4 cladding at elevated temperatures and its relevance to transient LWR fuel rod behaviour

In extensive out-of-pile experiments from 500 to 900° C it has been shown that, of all the volatile fission products in a LWR fuel rod, only iodine can cause low ductility failure of Zircaloy-4 tubing due to stress corrosion cracking up to about 800° C. The critical iodine concentration above which brittle cladding failure occurs was determined as a function of temperature in the absence and presence of UO₂ fuel. A comparison of these values with the amount expected in the fuel cladding gap during a LOCA transient shows that a clear influence of iodine on burst strain can be expected only up to 700° C. This is in agreement with the results of in-pile LOCA tests performed in the FR-2 reactor with high burnup fuel rods. Since the burst temperatures during a LOCA transient would generally be above 700° C, an influence of iodine on burst strain is not very probable in a LOCA. However, with respect to ATWS transients where the maximum cladding temperatures would be below 700° C, an influence of iodine on the mechanical properties of zircaloy can be expected.     

Examination of the chemical composition of irradiated zirconium based fuel claddings at the metal/oxide interface by TEM

Detailed post-irradiation examinations have been performed at PSI on three fuel rods with differing cladding materials revealing different corrosion behaviour. The rods had been irradiated for 3-5 cycles at Gösgen nuclear power plant (pressurised water reactor), Switzerland. As zirconium corrosion is proceeding at the metal/oxide interface, extended micro-structural analyses were performed by transmission electron microscopy (TEM), expecting to possibly reveal phenomena explaining the varying corrosion resistance. This paper reports on the distribution of oxygen at the metal/oxide interface examined by energy dispersive X-ray spectroscopy (EDS) in TEM, while other micro-structural investigations have been published earlier [1]. In order to get some statistical confidence in the analyses, three neighbouring TEM samples of each cladding variant were studied. The oxygen concentration profiles of the three alloys (i.e. low-tin Zircaloy-4, Zr2.5%Nb and extra low-tin (Sn 0.56%)) both in the oxide and metal close to the metal/oxide interface are compared. The results of the examinations show the composition of the oxide in the vicinity of the interface to be sub-stoichiometric for all three materials, indicating an oxide layer adjacent to the interface, with diffusion-controlled access of oxygen to the metal/oxide interface. The metallic parts show highest oxygen concentrations at the metal/oxide interface which are reduced towards the bulk metal, pointing towards the expected second diffusion-controlled process leading to α-Zr (O). Based on the experimental results values for the diffusion coefficients in the range of 0.8-6.0 × 10⁻²⁰ m² s⁻¹ are estimated for the oxygen dissolution process, the diffusion coefficient in Zircaloy-4 being six times higher than for the other two less corroding alloys. This finding is in contradiction with the present assumptions about the corrosion mechanism, and confirms the expected but not so far reported diffusion controlled oxidation of different zirconium alloys. It also points towards a corrosion rate that is at least partly governed by the diffusion coefficient of oxygen in metal that is different for different alloys, unlike what has been assumed till present.     

Developments in PWR cladding corrosion: materials and models

In pressurized water reactors Zircaloy-4 is a standard fuel cladding material. The aim of this paper is to present and evaluate corrosion data generated both in-reactor, and out-of-reactor on PWR claddings made of both Zircaloy-2 and Zircaloy-4 materials. The oxide thickness measurements of cladding tubes irradiated in the Ringhals 3 reactor, and oxide weight gain measurements carried out in Sandvik autoclaves at 400°C, 10.3 MPa clearly show that the stress relief annealed Zircaloy-2 is more corrosion resistant than Zircaloy-4 produced with an identical fabrication route. Furthermore, autoclave tests indicate that the hydrogen pickup fraction of the two alloys is very similar. The obtained data have been evaluated in regard to chemical composition and heat treatment. In addition, computer models, which simulate thermal and hydraulic reactor conditions and corrosion kinetic processes simultaneously, have been used to predict the in-reactor corrosion behaviour of the claddings.