Some fractographic aspects of hydrogen-induced delayed cracking in Zr–2.5 wt% Nb alloys

The fractographic features after hydrogen-induced delayed cracking (HIDC) in Zr-2.5 wt% Nb pressure tube material have been studied as a function of stress intensity factor (K), temperature and hydride dispersion. The important observations were of regularly spaced ductile striations parallel to the main crack front, and brittle cleavage markings between the striations. The striations were cusped due to dimple formation, and individual cleavage regions were separated by ductile ridges or shear cliffs. The size and morphology of the cleavage regions was consistent with the size and preferred orientation of hydrides near the crack tip. The inter-striation spacing increased exponentially with temperature, but within the range of study, was essentially independent of K. Neither hydride content nor dispersion had a significant effect on the fracture morphology. The results are consistent with a discontinuous mechanism of crack propagation involving the accumulation of hydride at the crack tip, followed by fracture through this embrittled region and subsequent crack arrest. The observations support recent theoretical models for HIDC which interpret the inter-striation spacing as a critical hydride length for crack advance.     

Rupture of spent fuel Zircaloy cladding in dry storage due to delayed hydride cracking

Delayed hydride cracking in the Zircaloy alloy has been considered as a possible degradation mechanism of spent nuclear fuel cladding in interim dry storage. Some recent in-core fuel failures indicated that a long axial crack developed in the cladding was a secondary failure by delayed hydride cracking. The aim of this study is to define the effects of hydride reorientation on the failure of Zircaloy cladding. Different hydride orientations, the amount of zirconium hydride and various cracking types, all have been considered for their effects on the crack growth and stability of the cladding, and have been thoroughly discussed in this paper. A finite element computer code, ANSYS, has been used in conjunction with the strain energy density theory. In summary, the crack propagation will be aggravated if the hydride orientation is shifted from the circumferential to the radial direction. For a larger crack length, the zirconium hydride plays an important role in affecting the crack growth because the strain energy density factor increases as the hydride approaches the crack tip. Furthermore, when thermal effects are considered, a compressive stress exists at the inner side of the cladding, while a tensile stress is found at the outer side of cladding, thus resulting in crack propagation from the outer side to the inner side of the cladding. These findings are in accordance with other experimental results in related literature.     

Influence of Cladding-Peripheral Hydride on Mechanical Fuel Failure under Reactivity-Initiated Accident Conditions

In order to investigate the influence of hydrogen embrittlement on fuel failure under reactivity-initiated accident (RIA) conditions, pulse irradiation experiments were performed with unirradiated fuel rods at the Nuclear Safety Research Reactor (NSRR). Fresh cladding was pre-hydrided so that the other factors of cladding degradation, such as irradiation damage and oxidation, were excluded. Hydride clusters are circumferentially oriented and localized in the cladding periphery in order to simulate ‘hydride rim’ which is formed in high burnup PWR cladding. The present study demonstrated hydride-assisted pellet-cladding mechanical interaction (PCMI) failure which has been observed in high burnup fuel experiments. The fuel enthalpy at failure was lower when the cladding had a thicker hydride rim where surface cracks were easily generated. It indicates that the failure limit is highly correlated with the stress intensity factor assuming that the crack depth is equivalent to the hydride rim thickness. Hence, we conclude that hydride rim formation is the primary factor of decreasing the failure limit for high burnup fuels. Based on the experimental results together with an analysis on cladding mechanical state during PCMI, the present study suggests a process of through-wall crack generation which is originated with brittle cracking within the hydride rim.     

CANDU及RBMK压力管锆合金的氢致延迟断裂研究

采用紧凑拉伸试样(CT），在恒定载荷、不同氢含量、不同温度条件下，测量了CANDU堆和RBMK堆Zr-2．5Nb压力管材料氢致延迟开裂速率。用金相显微镜和扫描电镜观察断口及氢化物形貌，并测量临界应力场强度因子及开裂速率，对材料的微结构及氢化物分布进行分析。结果表明，氢致延迟断裂(DHC)生长呈阶梯状。与CANDU压力管比较，RBMK压力管的DHC开裂速率将近低一个数量级。其原因是RBMK压力管的屈服强度比CANDU压力管低得多。     

Crack growth pattern and threshold stress intensity factor, KIH, of Zr–2.5Nb alloy with the notch direction

The objective of this study is to obtain a better understanding of the threshold stress intensity factor for an initiation of delayed hydride cracking (DHC) in a Zr–2.5Nb pressure tube. By changing the crack propagation from the longitudinal direction to the circumferential direction, the threshold stress intensity factor, K_IH, and the crack growth pattern were investigated in the Zr–2.5Nb pressure tube with a strong circumferential texture. The threshold stress intensity factor, K_IH, was discussed phenomenologically based on the crack growth pattern and analytically as a function of the tilting angle of hydride habit planes to the cracking plane. A supplementary experiment was conducted to demonstrate a linear decrease of K_IH with an increase in the basal pole component in the cracking plane. Thus, it is concluded that the DHC is controlled by the nucleation and growth of the hydride precipitates on the habit plane.     

Fatigue Life and Crack Growth Mechanisms of the Type 316LN Austenitic Stainless Steel in 310°C Deoxygenated Water

The low cycle fatigue tests of the type 316LN stainless steel were conducted to investigate the cracking mechanisms in high-temperature water. The fatigue lives of the specimens tested in 310°C deoxygenated water were considerably shorter than those tested in air. For the specimens tested in 310°C deoxygenated water, the evidences for the metal dissolution such as the stream downed feature, the blunt crack shape, and the wider crack opening were observed but rather weakly. In the same specimens, the evidences for the hydrogen-induced cracking such as the coalescence of microvoids and the decrease of the dislocation spacing at the crack tip were observed rather clearly. Therefore, it is thought that the hydrogen-induced cracking is mainly responsible for the reduction in the fatigue life of the type 316LN stainless steel in 310°C deoxygenated water while the effect of metal dissolution is less significant. The hydrogen-induced cracking is more pronounced in the slower strain rates. This behavior is in accordance with the larger reduction in the fatigue life at the slower strain rates. Furthermore, the fatigue life and the dislocation spacing show the minimum value in the strain rate range from 0.008 to 0.04%/s, which indicates the existence of the critical strain rate.     

Delayed hydride cracking of spent fuel rods in dry storage

Failures of zirconium alloy cladding tubes during a long-term storage at room temperature were first reported by Simpson and Ells in 1974, which remains unresolved by the old delayed hydride cracking (DHC) models. Using our new DHC model, we examined failures of cladding tubes after their storage at room temperature. Stress-induced hydride phase transformation from γ to δ at a crack tip creates a difference in hydrogen concentration between the bulk region and the crack tip due to a higher hydrogen solubility of the γ-hydride, which is a driving force for DHC at low temperatures. Accounting for our new DHC model and the failures of zirconium alloy cladding tubes during long-term storage at room temperature, we suggest that the spent fuel rods to be stored either in an isothermal condition or in a slow cooling condition would fail by DHC during their dry storage upon cooling to below 180 °C. Further works are recommended to establish DHC failure criterion for the spent fuel rods that are being stored in dry storage.     

The first step for delayed hydride cracking in zirconium alloys

Two models for delayed hydride cracking (DHC) in zirconium alloys are distinguished by their first step:

-

The loading of a crack induces hydride precipitation. The hydride is postulated to create a hydrogen concentration gradient, where the bulk concentration is greater than that at the crack tip. This concentration gradient is taken as the driving force for diffusion of hydrogen to the crack tip, and subsequent hydride growth. This model is called the precipitate first model (PFM).

-

The tensile stress at the crack tip induces a gradient in chemical potential that promotes the diffusion of hydrogen to the crack tip. Hydrides form if the hydrogen concentration reaches the solubility limit for hydride precipitation. The mechanism is postulated to create a hydrogen concentration gradient, where the bulk concentration is lower than that at the crack tip. The gradient in chemical potential is taken as the driving force for diffusion of hydrogen to the crack tip, and subsequent hydride growth. This model is called the diffusion first model (DFM).

The second model, DFM, is developed. This model is shown to describe the main features of the experimental observations of DHC, without invoking new phenomena, such as reduction in the solubility limit for precipitation of hydride, as required by the PFM.     

Comment on ‘The first step for delayed hydride cracking in zirconium alloys’ by G.A. McRae et al., J. Nucl. Mater. 396 (2010) 130-143

The aim of this paper is a reply to McRae et al.’s paper entitled “The first step for delayed hydride cracking (DHC) in zirconium alloys” claiming that the first step of DHC is hydrogen diffusion, not nucleation of hydrides as demonstrated by Kim’s new model. Despite the authors’ claim that the crack tip concentration is higher than the bulk concentration due to the stress gradient, their claim violates the thermodynamic principle that the stressed region should have a lower potential of hydrogen or lower hydrogen solubility than the unstressed region. Furthermore, it is demonstrated that the Diffusion First Model (DFM) proposed by the author is defective in terms of kinetics because hydrogen diffusion before hydride nucleation just governs the rate of hydride nucleation, neither the rate of hydride growth nor the crack growth rate (CGR).     

国产316LN不锈钢在模拟AP1000一回路水环境中的疲劳行为

一回路水环境下的疲劳性能是核电站主管道设计寿命评估的重要参数。针对国产主管道材料316LN开展了模拟AP1000一回路水环境的低周疲劳试验,分析了疲劳行为和失效机理。研究结果表明:国产316LN峰值应力随应变幅的增大而增大,大应变幅试样在疲劳过程中先后发生了循环硬化、循环软化和失稳,而小应变幅试样在失稳前未发生明显的循环硬化和循环软化;在应变幅由0.2%逐渐增加至1.2%的过程中,疲劳周次从105逐渐降低至102;疲劳断口具有典型的疲劳断口特征,裂纹萌生于试样表面,以穿晶方式垂直于主应力方向扩展,裂纹扩展区具有典型的疲劳辉纹,辉纹上有菱形颗粒状腐蚀产物,环境辅助开裂机制倾向于氢致开裂。     

Advanced TEM characterization of stress corrosion cracking of Alloy 600 in pressurized water reactor primary water environment

Advanced transmission electron microscopy techniques were carried out in order to investigate stress corrosion cracking in Alloy 600 U-bend samples exposed in simulated PWR primary water at 330 °C. Using high-resolution imaging and fine-probe chemical analysis methods, ultrafine size oxides present inside cracks and intergranular attacks were nanoscale characterized. Results revealed predominance of Cr₂O₃ oxide and Ni-rich metal zones at the majority of encountered crack tip areas and at leading edge of intergranular attacks. However, NiO-structure oxide was predominant far from crack tip zones and within cracks propagating along twin boundaries and inside grains. These observations permit to suggest a mechanism for intergranular stress corrosion cracking of Alloy 600 in PWR primary water. Indeed, the results suggest that stress corrosion cracking is depending on chromium oxide growth in the grain boundary. Oxide growth seems to be dependent on oxygen diffusion in porous oxide and chromium diffusion in strained alloy and in grain boundary beyond crack tip. Strain could promote transport kinetic and oxide formation by increasing defaults rate like dislocations.     

Delayed hydride cracking in Zr-2.5Nb tube with the cooling rate and the notch tip shape

The objective of this study is to demonstrate the feasibility of the Kim’s delayed hydride cracking (DHC) model. To this end, this study has investigated the velocity and incubation time of delayed hydride cracking (DHC) for the water-quenched and furnace-cooled Zr-2.5Nb tubes with a different radius of notch tip. DHC tests were carried out at constant K_I of 20 MPa √m on cantilever beam (CB) specimens subjected to furnace cooling or water quenching after electrolytic charging with hydrogen. An acoustic emission sensor was used to detect the incubation time taken before the start of DHC. The shape of the notch tip changed from fatigue cracks to smooth cracks with its tip radius ranging from 0.1 to 0.15 mm. The DHC incubation time increased remarkably with the increased radius of the notch tip, which appeared more strikingly on the furnace-cooled CB specimens than on the water-quenched ones. However, both furnace-cooled and water-quenched CB specimens indicated little change in DHC velocity with the radius of the notch tip unless their notch tip exceeded 0.125 mm. These results demonstrate that the nucleation rate of hydrides at the notch tip determines the incubation time and the DHC velocity becomes constant after the concentration of hydrogen at the notch tip reaches terminal solid solubility for dissolution (TSSD), which agrees well with the Kim’s DHC model. A difference in the incubation time and the DHC velocity between the furnace-cooled and water-quenched specimens is attributed to the nucleation rate of reoriented hydrides at the notch tip and the resulting concentration gradient of hydrogen between the notch tip and the bulk region.     

Formation and growth of cracking in feed water pipes and RPV nozzles

The observation of numerous small and large cracks in ferritic feed water pipes of boiling (BWR) and pressurized water reactors (PWR) in the last few years has led to basic research into the causes of cracking and the crack growth mechanisms.In horizontal feed water pipe sections connected to nozzles of reactor pressure vessels (RPV) of BWR's as well as of steam generators (SG) of PWR's, circumferential macro and micro cracks were detected. These cracking phenomena could be observed in base material of pipes as well as in weld seam regions. The examination of the stress state displayed that the cracked pipe regions have been exposed to a number of cyclic thermal transients (thermal shock, flow stratification) during start-up (hot stand-by) and shut-down periods of the plants. During thermal transient periods, local and global cyclic stresses in the referred pipe cross sections have been induced which in interaction with the influence from environment (in operation as well as in shut-down periods) and local geometrical imperfections led to the initiation and formation of macro and micro cracks.In the reactor water clean-up system of BWR through which reactor water is fed from the RPV to the main feed water line, two longitudinally welded elbows have been detected to be severely cracked. Both elbows have been subjected to an internal pressure corresponding to RPV and additionally to a relevant “in-plane” bending moment. These longitudinal cracks were found to be started from the inner elbow surface. In one case the longitudinal crack was situated in the base material and was enlarged to leakage. In the second elbow the longitudinal crack was located in the heat affected zone (HAZ) of a longitudinal weld. In both cases the macro cracks started either from corrosion pits located in defective areas of the magnetic protection layer or from geometrical notches (weld root). The semi-elliptic small cracks got linked to more extended shallow cracks.Formation and growth mechanism of these cracks have been studied at the MPA Stuttgart in laboratory under simulated operation conditions which were held as realistic as possible compared with those in nuclear power plants.The results of experimental studies in laboratory as well as conclusions based on the above mentioned cracking phenomena in piping have been used as basic information for a realistic design of large scale (RPV) thermal shock experiments under operation conditions. The formation and growth mechanism of these cracks and their detection by means of NDE during thermal transients at the inner surface of RPV nozzle and at the adjacent cylindrical areas of RPV shell will be described.     

Research Program to Elucidate Outside-in Failure of High Burnup Fuel Cladding

Among a series of power ramp tests on 25 Zr-lined segment rods of burnup ranging from 43 to 61 GWd/t, five segment rods failed during the power ramp tests. One segment rod irradiated for 3 cycles (43 GWd/t) failed with a pinhole due to PCI/SCC. The rest of higher burnups failed with an axial crack on the outer surface. The failure threshold power tended to decrease as burnup increases.

Post irradiation examinations revealed increased cladding hydrogen absorption and its precipitates in the cladding outer rim after 4 and 5 cycle irradiations, in contrast to a uniform hydride distribution and a small hydrogen content after 3 cycle irradiation. Metallographic observations suggested an axial crack failure mode induced by the combined effects of high stress and hydrides precipitated in a radial direction during power ramp.

The axial crack failure during the power ramp is supposed to be initiated by a cracking of radial hydride formed by hydride re-distribution and re-orientation at the cladding outer rim and to propagate through a process of hydride concentration and precipitation at the crack tip. Research programs of experimental and analytical studies to clarify the conditions of such mechanism are on-going focusing on the hydrogen behavior and mechanical performance of the irradiated cladding.     

Review of the thermodynamic basis for models of delayed hydride cracking rate in zirconium alloys

A review is given of the thermodynamic basis of a model developed by Dutton and Puls for the rate of subcritical crack propagation by delayed hydride cracking in zirconium alloys. This review was prompted, in part, by the publications of a series of recent papers by Kim and co-workers in which it is claimed that the thermodynamic basis of the Dutton and Puls model and its subsequent refinements is incorrect, prompting them to propose a new model. This review demonstrates the validity of the original model and shows the origin of the error made by Kim in claiming that the Dutton and Puls model was incorrectly formulated. It also explains the reasons why Kim’s new delayed hydride cracking model is incorrect. This review was further prompted by the author’s realization that the series of papers documenting the development of the various versions of the original Dutton and Puls model contain typographical errors, differences in sign convention, differences in input data, minor errors and/or changes in formal representation as well as occasional misleading, confusing or incorrect statements of the physical significance of the thermodynamic basis of the model. All of these shortcomings could have resulted in misunderstandings regarding the correct formulation of the model and the physical significance of the results. Therefore another important purpose of this review is to provide an updated treatment of the original version that puts all subsequent versions of the DHC model on a consistent thermodynamic basis.     

An elastoplastic phase-field model for the evolution of hydride precipitation in zirconium. Part II: Specimen with flaws

An elastoplastic phase-field model, described in Part I, was applied to bulk materials containing flaws such as sharp cracks and blunt notches. An additional set of long range order parameters, namely, stress-free strains for flaws, was introduced. The nucleation and growth of hydrides near a void or a crack were simulated by the proposed elastoplastic phase-field model. The effects of notch root radius, hydrogen concentration in solid solution, yield stress of the matrix and the level of externally applied stress on hydride morphology around flaws were studied. It is demonstrated that parameters such as the distribution of the tensile stress component perpendicular to the hydride platelet normal may be closely monitored during hydride growth near a flaw with or without externally applied stresses. Combined with a fracture criterion and real experimental data, the model is capable of predicting the rate and morphology of hydride precipitation, and crack initiation near flaws.     

Primary water stress corrosion cracking behaviors in the shot-peened Alloy 600 TT steam generator tubings

The effect of compressive residual stress on the primary water stress corrosion cracking behavior was investigated, based on the J-1 and J-2 nuclear power plant data. The following analyses were performed such as: (i) Weibull slope; (ii) crack growth rate; (iii) average crack length; (iv) crack length distribution. Alloy 600 TT exhibits strong heat to heat variations in its sensitivity to PWSCC. Crack growth rate was retarded after shot-peening. The compressive residual stress induced by shot-peening was more effective on new, short cracks, than on existing, long cracks. However, whether the ‘new’ cracks were initiated after peening is an unresolved issue, due to the present ECT sensitivity limit.     

A risk-informed approach to assessment of DHC initiation in pressure tubes of CANDU reactors

The delayed hydride cracking (DHC) of flaws and cracks in pressure tubes is a serious form of degradation in the reactor core. CSA standard N285.8 (2005) recommends deterministic and probabilistic procedures for assessing the potential of DHC initiation from flaws that are generated by fretting or any other mechanism. Although the deterministic method is simple, it lacks a quantitative risk-informed basis for the assessment. On the other hand, a full probabilistic method based on simulation is tedious to implement. This paper presents an efficient, reliability-based approach that bridges the gap between a simple deterministic analysis and full Monte Carlo simulations. In the proposed method, a deterministic DHC initiation criterion is calibrated to specified target probability levels. The main advantage of the proposed approach is that it provides a practical, risk-informed basis for DHC initiation assessment while retaining the simplicity of the deterministic method.     

Modelling the early development of secondary side stress corrosion cracks in steam generator tubes using incomplete random tessellation

A thorough understanding of the secondary side stress corrosion cracking of Inconel 600 in steam generator (SG) tubes seems to be still somewhat in the future. Especially the early phase of the development of cracks, also called the initiation phase, is beyond the present state-of-the-art explanations. An effort was, therefore, made to propose modelling and visualisation of the kinetics of secondary side stress corrosion crack initiation and growth on the grain-size scale:

An incomplete random tessellation is used to approximate the random planar grain structure.

The crack initiation is modelled by random processes, taking into account the most important factors such as proximity of the aggressive medium and the orientation of the grain boundaries relative to the stress field.

The stochastic process describing crack growth accounts for crack branching, coalescence and interference between neighbouring cracks.

Several numerical examples are provided to demonstrate the versatility of the proposed method. Reasonable qualitative agreement with metallographic results is shown.     

A probabilistic method for leak-before-break analysis of CANDU reactor pressure tubes

In modern CANDU nuclear generating stations, pressure tubes of cold-worked Zr-2.5Nb material are used in the reactor core to contain the fuel bundles and the heavy water (D₂O) coolant. The pressure tubes operate at an internal pressure of 10 MPa and temperatures ranging from 250°C at the inlet to 310°C at the outlet. Over the expected 30 year lifetime of these tubes, they would be subjected to a total fluence of 3×10²⁶ n m⁻². In addition, these tubes gradually pick up deuterium as a result of a slow corrosion process. When the hydrogen plus deuterium concentration in the tubes exceeds the hydrogen/deuterium solvus, the tubes are susceptible to a crack initiation and propagation process called delayed hydride cracking (DHC). If undetected, such a cracking mechanism could lead to unstable rupture of the pressure tube. The service life of the pressure tubes is determined, in part, by changes in the probability for the rupture of a tube. This probability is made up of the probability for crack initiation by DHC multiplied by the sum of the probabilities of break-before-leak and leak-before-break (LBB). A probabilistic model, BLOOM, is described which makes it possible to estimate the cumulative probabilities of break-before-leak and LBB. The probability of break-before-leak depends on the crack length at first leak detection and the critical crack length. The probability of a LBB depends on the shut-down scenario used. The probabilistic approach is described in relation to an example of a possible shut-down scenario. Key physical input parameters into this analysis are pressure tube mechanical properties, such as the crack length at first coolant leakage, the DHC velocity and the critical crack length. Since none of these parameters are known precisely, either because they depend on material properties, which vary within and between pressure tubes, and/or because of measurement errors, they are given in terms of their means and standard deviations at the different temperatures and pressures defined by the shut-down scenario.