Impact of localized deformation on IASCC in austenitic stainless steels

Localized deformation has been identified as a potential primary contributor to IASCC. Seven austenitic alloys were irradiated to 1 and 5 dpa at 360 °C using 2-3.2 MeV protons and were tested both in simulated BWR environment and in argon. Cracking susceptibility was evaluated at both 1% and 3% strain intervals using crack length per unit area. Stacking fault energy (SFE), hardness, radiation-induced segregation (RIS) and localized deformation were characterized and their correlations with cracking were evaluated using a proposed term, correlation strength. Both SFE and hardness contributed to cracking but neither was the dominant factor. RIS did not play an important role in this study. The correlation strength of localized deformation with IASCC was found to be significantly higher than for others parameters, implying that localized deformation is the most important factor in IASCC. Although not well understood, localized deformation may promote cracking through intensive interaction of dislocations in slip channels with grain boundaries.     

Modelling of the effect of dislocation channel on intergranular microcrack nucleation in pre-irradiated austenitic stainless steels during low strain rate tensile loading

In the present article, the effect of dislocation channel on intergranular microcrack nucleation during the tensile deformation of pre-irradiated austenitic stainless steels is studied. Because several slip planes are activated within the dislocation channel, the simple dislocation pile-up model seems not well suited to predict grain boundary stress field. Finite element computations, using crystal plasticity laws and meshes including a channel of finite thickness, are also performed in order to study the effect of some microstructural characteristics on grain boundary stress field. Numerical results show that: the thickness and the length of the dislocation channel influence strongly the grain boundary normal stress field. The grain boundary orientation with respect the stress axis does not affect so much the grain boundary normal stresses close to the dislocation channel. On the contrary far away the dislocation channel, the grain boundary stress field depends on the grain boundary orientation. Based on these numerical results, an analytical model is proposed to predict grain boundary stress fields. It is valuable for large ranges of dislocation channel thickness, length as well as applied stress. Then, a macroscopic microcrack nucleation criterion is deduced based on the elastic-brittle Griffith model. The proposed criterion predicts correctly the influence of grain boundary characteristics (low-angle boundaries (LABs), non-coincident site lattice (non-CSL) high-angle boundaries (HABs), special grain boundaries (GBs)) on intergranular microcrack nucleation and the macroscopic tensile stress required for grain boundary microcrack nucleation for pre-irradiated austenitic stainless steels deformed in argon environment. The criterion based on a dislocation pile-up model (Smith and Barnby) underestimates strongly the nucleation stress. These results confirm that pile-up models are not well suited to predict microcrack nucleation stress in the case of dislocation channels impacting grain boundary. The proposed criterion is applied to the prediction of the IASCC macroscopic nucleation stress for pre-irradiated material tested in PWR environment and the predictions are discussed with respect to experimental data. Finally, the limitations of the continuum modelling are discussed.     

Analysis of tensile deformation and failure in austenitic stainless steels: Part I - Temperature dependence

This paper describes the temperature dependence of deformation and failure behaviors in the austenitic stainless steels (annealed 304, 316, 316LN, and 20% cold-worked 316LN) in terms of equivalent true stress-true strain curves. The true stress-true strain curves up to the final fracture were calculated from tensile test data obtained at −150 to 450 °C using an iterative finite element method. Analysis was largely focused on the necking and fracture: key parameters such as the strain hardening rate, equivalent fracture stress, fracture strain, and tensile fracture energy were evaluated, and their temperature dependencies were investigated. It was shown that a significantly high strain hardening rate was retained during unstable deformation although overall strain hardening rate beyond the onset of necking was lower than that of the uniform deformation. The fracture stress and energy decreased with temperature up to 200 °C and were nearly saturated as the temperature came close to the maximum test temperature 450 °C. The fracture strain had a maximum at −50 to 20 °C before decreasing with temperature. It was explained that these temperature dependencies of fracture properties were associated with a change in the dominant strain hardening mechanism with test temperature. Also, it was seen that the pre-straining of material has little effect on the strain hardening rate during necking deformation and on fracture properties.     

Analysis of tensile deformation and failure in austenitic stainless steels: Part II - Irradiation dose dependence

Irradiation effects on the stable and unstable deformation and fracture behavior of austenitic stainless steels (SSs) have been studied in detail based on the equivalent true stress versus true strain curves. An iterative finite element simulation technique was used to obtain the equivalent true stress-true strain data from experimental tensile curves. The simulation result showed that the austenitic stainless steels retained high strain hardening rate during unstable deformation even after significant irradiation. The strain hardening rate was independent of irradiation dose up to the initiation of a localized necking. Similarly, the equivalent fracture stress was nearly independent of dose before the damage (embrittlement) mechanism changed. The fracture strain and tensile fracture energy decreased with dose mostly in the low dose range <∼2 dpa and reached nearly saturation values at higher doses. It was also found that the fracture properties for EC316LN SS were less sensitive to irradiation than those for 316 SS, although their uniform tensile properties showed almost the same dose dependencies. It was confirmed that the dose dependence of tensile fracture properties evaluated by the linear approximation model for nominal stress was accurate enough for practical use without elaborate calculations.     

The role of irradiated microstructure in the localized deformation of austenitic stainless steels

Localized deformation has emerged as a potential factor in irradiation-assisted stress corrosion cracking of austenitic stainless steels in LWR environments and the irradiated microstructure may be a critical factor in controlling the degree of localized deformation. Seven austenitic alloys with various compositions were irradiated using 2-3 MeV protons to doses of 1 and 5 dpa at 360 °C. The irradiated microstructure consisting of dislocation loops and voids was characterized using transmission electron microscopy. The degree of localized deformation was characterized using atomic force microscopy on the deformed samples after conducting constant extension rate tension tests to 1% and 3% strain in argon. Localized deformation was found to be dependent on the irradiated microstructure and to correlate with hardening originating from dislocation loops. Dislocation loops enhance the formation of dislocation channels and localize deformation into existing channels. On the contrast, voids mitigate the degree of localized deformation. The degree of localized deformation decreases with SFE with the exception of alloy B. Localized deformation was found to have similar dependence on SFE as loop density suggesting that SFE affects localized deformation by altering irradiated microstructure.     

Helium release in uranium dioxide in relation to grain boundaries and free surfaces

Nuclear reaction analyses (NRA) based on the ³He(²H,⁴He)¹H reaction were previously performed to follow the evolution of implanted ³He in polycrystalline UO₂ samples. Experimental results pointed to an enhancement above 800 °C of the diffusion coefficient of helium over several microns in the vicinity of the grain boundaries, with respect to the diffusion coefficient within the grain. This was ascribed to the fact that grain boundaries are probably defect sinks which locally modify the defect concentrations.This study aims at demonstrating the particular effect of grain boundaries on helium migration. To this end, ³He implanted polycrystalline UO₂ samples were cracked then annealed at 900 °C. Helium migration in the vicinity of the grain boundaries and near the crack was investigated by means of NRA microanalyses. Helium depletion extends over far larger distances in the vicinity of the grain boundaries than near the crack. Experimental evidence has been collected of the particular effect of grain boundaries on helium migration, which do not act as free surfaces at which helium atoms are simply released.     

Investigation on the formation of serrated grain boundaries with grain boundary characteristics in an AISI 316 stainless steel

The formation of serrated grain boundaries (GBs) depending on the GB characteristics has been investigated by using an electron backscattered diffraction (EBSD) technique and a transmission electron microscopy (TEM) in an AISI 316 stainless steel. It was observed that at the early stage of aging treatment, the GB morphology was changed from flat to wavy at random GBs without any indication of M₂₃C₆ carbide formation, and no GB serration at special GBs (lower than Σ29) was found. The comparison study on the misorientation angle between two neighboring grains indicated that the occurrence of GB serration at random GBs is attributed to the reduction of the total GB energy. Random GBs with high energy tend to be serrated, resulting in the formation of two segments with lower energies. On the other hands, the special GBs may be less likely to form serrated GBs due to their lower GB energy.     

Localized deformation as a key precursor to initiation of intergranular stress corrosion cracking of austenitic stainless steels employed in nuclear power plants

Cold-work has been associated with the occurrence of intergranular cracking of stainless steels employed in light water reactors. This study examined the deformation behavior of AISI 304, AISI 347 and a higher stacking fault energy model alloy subjected to bulk cold-work and (for 347) surface deformation. Deformation microstructures of the materials were examined and correlated with their particular mechanical response under different conditions of temperature, strain rate and degree of prior cold-work. Select slow-strain rate tensile tests in autoclaves enabled the role of local strain heterogeneity in crack initiation in pressurized water reactor environments to be considered. The high stacking fault energy material exhibited uniform strain hardening, even at sub-zero temperatures, while the commercial stainless steels showed significant heterogeneity in their strain response. Surface treatments introduced local cold-work, which had a clear effect on the surface roughness and hardness, and on near-surface residual stress profiles. Autoclave tests led to transgranular surface cracking for a circumferentially ground surface, and intergranular crack initiation for a polished surface.     

Development of thermal fatigue testing apparatus with BWR water environment and thermal fatigue strength of austenitic stainless steels

A thermal fatigue testing apparatus was developed in order to clarify the fatigue behavior in BWR environment. Pressurized high and low temperature pure water were alternately supplied into an autoclave with a small cylindrical specimen. Then a fatigue specimen was subjected to homogeneous thermal stress through the wall thickness. Fatigue crack initiation behavior was observed with the replication method and compared with the mechanical fatigue strength performed in air and high temperature water. The thermal fatigue strength of type 304 and 316 nuclear grade (316NG) stainless steels agreed closely with the mechanical fatigue strength, when transforming the nominal stress amplitude to the fictitious stress amplitude by using the mean value of strain amplitudes for room temperature and 288°C.     

Correlation between composition of passive layer and corrosion behavior of high Si-containing austenitic stainless steels in nitric acid

Austenitic stainless steels with 18% Cr have a good corrosion behavior in pure nitric acid. However, when oxidizing power of the solution increases, this kind of stainless steels faces a severe intergranular corrosion. Adding a sufficiently high concentration of silicon to the steel avoids this type of corrosion: in oxidizing solutions, those stainless steels exhibit generalized corrosion but their dissolution rate is higher than the one of stainless steels without silicon. To find out the role of silicon on such effects, the corrosion behavior of two different stainless steels with equivalent chromium content but with different silicon content (304L steel and Uranus S1N) has been studied in concentrated nitric acid solutions. Correlations have been evidenced between the passive layer composition investigated by XPS analysis and the corrosion behavior characterized by electrochemical techniques. The presence of silicon in the steel changes neither the oxidation state of chromium or iron, nor the ratio between iron and chromium in the passive layer. Silicon is present in the passive layer in an important content (35 at.%) and thus decreases the chromium content of the passive layer (80 and 50 at.% respectively for 304L steel and Uranus S1N after nitric passivation). Uranus S1N exhibits a less protective passive layer and so its generalized corrosion rate is higher than the one of 304L steel. A selective deposition of platinoïds highlights differences of polarization distribution on the surface between the grain boundaries and grain faces for theses steels. For Uranus S1N, the similar electrochemical behavior of grain boundaries and faces might be connected with the homogeneous silicon distribution.     

The effects of temperature and dose-rate variations on the creep of austenitic stainless steels in the dounreay fast reactor

Loaded helices of cold-worked, austenitic stainless steels (En58B, E, J, FV548 and type 316) have been irradiated in DFR at damage rates ranging over an order of magnitude at temperatures between 513 and 633 K. Creep rates varied between steels but a common pattern emerged where strain per unit dose appeared to increase with decreasing dose-rate. Comparison with experiments at 773 K on three of the steels reveals creep rate increases by factors ranging from 1.2 to 5. The results are discussed in the context of other fast reactor data and possible explanations of the findings are considered.     

Effects of nitrogen on low-cycle fatigue properties of type 304L austenitic stainless steels tested with and without tensile strain hold

A quantitative analysis of the effects of nitrogen on high temperature low-cycle fatigue without and with tensile strain hold at 600 °C has been conducted for type 304L stainless steels. For better understanding of the role of nitrogen on grain boundary precipitation, the grain boundary segregation of nitrogen was analyzed by Auger electron spectroscopy. The nitrogen addition is found to give relatively better resistance to creep-fatigue than continuous low-cycle fatigue. This in turn improves the fatigue life. This is due to the retardation of the precipitation of carbides at the grain boundary and reduction in the density of grain boundary cavitation sites which are the main factor of grain boundary damage under creep-fatigue test.     

High temperature creep and cyclic deformation behaviour of AISI 316 L(N) austenitic steel and its modelling with unified constitutive equations

Creep tests at constant stress and low cycle fatigue (LCF) tests were performed with a view to investigating and modelling the deformation behaviour of AISI 316 L(N) austenitic stainless steel at 700 °C. All experiments were done on samples taken from two different sheets of the same batch of material.The creep stresses were selected from the high stress range. The results obtained from creep tests on samples from different sheets are compared with each other. The differences between them and the results of a creep test carried out at constant load are indicated.The LCF experiments were strain controlled. The effects of strain rate and strain amplitude on the cyclic hardening behaviour were investigated.The parameters of a set of constitutive equations are determined from these data. The quality of the parameter fit is discussed.     

The use of model updating for reliable finite element modelling and fault diagnosis of structural components used in nuclear plants

Reliable finite element (FE) modelling in structural dynamics is very important for studies related to the safety of structural components used in the nuclear power industry. FE model updating is a tool to produce these reliable models. The method uses an initial FE model and experimental modal data of the structural components to modify physical parameters of the initial FE model, and a number of approaches have been developed to perform this task. This paper presents an overview of model updating and its use in fault diagnosis, using typical examples. The paper concentrates on the usefulness of the updating method, rather than describing the different updating methods in detail.     

Comparison of experimental and finite element analytical results for the strength and the deformation of pipes with local wall thinning subjected to bending moment

Fracture behaviors of pipes with local wall thinning are very important for the integrity of power plant piping system. In this study, monotonic bending tests without internal pressure are conducted on 48.6 mm diameter Schedule 80 (thickness, 5.1 mm) STS370 full-scale carbon steel pipe specimens. Fracture strengths of locally wall-thinned pipes were calculated by elasto-plastic analysis using finite element method. The elasto-plastic analysis was performed by FE code ANSYS. We simulated various types of local wall thinning that can be occurred at pipe surface due to coolant flow. Locally wall thinned shapes were machined to be different in size along the circumferential or axial direction of straight pipes. We investigated fracture strengths and failure modes of locally wall thinned pipes by four point bending test. And, the allowable limit of pipes with local wall thinning was investigated. In addition, we compared the simulated results by finite element analysis with experimental data. The failure mode, fracture strength and fracture behavior obtained from FE analyses showed well agreement with experimental results. From the test results, we identified three types of failure modes into ovalization, local buckling and crack initiation. These failure modes could be classified according to thinned depth, thinned length and thinned angle of a pipe. For locally wall-thinned specimens, maximum moments (M_max) were estimated by using the net-section stress criterion. Pipes with local wall thinning can be estimated using σ_u instead of σ_f because of 1.19σ_f  σ_u. Also, the axial strain affects failure modes occurred on local wall thinning. the allowable limit of local wall thinning for carbon steel pipe used can be given as follows; in the case of M_max ≥ M_y, if 10 ≤ l < 25 mm, d/t can be allowed to about 55%, and if 25 ≤ l < 100 mm, d/t can be allowed to about 50%. Also, if 100 ≤ l ≤ 120 mm, d/t can be allowed to about 29%.     

Boundary element method in nonlinear transient heat transfer of reactor solids with convection and radiation on surfaces

This paper deals with the boundary (integral) element method for non-steady conduction problems of solids, subject to non-linear convective and radiation conditions on surfaces. Boundary integral equations for the mixed-type and non-linear boundary conditions, both for the case with constant and variable heat conductivity are derived, modelled by non-conforming boundary elements, while domain integrals are evaluated within triangular cells. A test case is included to illustrate the described procedure.     

Analyzing the effect of displacement rate on radiation-induced segregation in 304 and 316 stainless steels by examining irradiated EBR-II components and samples irradiated with protons

Recent studies have indicated that, at temperatures relevant to fast reactors and light water reactors, void swelling in austenitic alloys progresses more rapidly when the radiation dose rate is lower. A similar dependency between radiation-induced segregation (RIS) and dose rate is theoretically predicted for pure materials and might also be true in complex engineering alloys. Radiation-induced segregation was measured on 304 and 316 stainless steel, irradiated in the EBR-II reactor at temperatures near 375 °C, to determine if the segregation is a strong function of damage rate. The data taken from samples irradiated in EBR-II is also compared to RIS data generated using proton radiation. Although the operational histories of the reactor irradiated samples are complex, making definitive conclusions difficult, the preponderance of the evidence indicates that radiation-induced segregation in 304 and 316 stainless steels is greater at lower displacement rate.     

Air contamination effects on the compatibility of liquid lithium with molybdenum; TZM,niobium, stainless steels,nickel and hastelloy N in stainless-steel vessels at 600°c

The effects of nonmetallic impurities on the compatibility of liquid lithium with molybdenum, TZM, niobium, type 304 and type 316 stainless steels, nickel and Hastelloy N were investigated. Three compatibility tests (test I, test II and test III), classified by the grade of air contamination of the lithium, were conducted at 600°C for about 1000 h in stainless-steel vessels. In each test the above-mentioned specimens were immersed together in the lithium. In test I weight gain was observed for all the specimens except nickel and Hastelloy N. However, in test II and test III, weight loss was observed for all the specimens. MoNi₃ was produced on the surface of the molybdenum and TZM specimens as a result of the reaction between molybdenum and nickel dissolved in the liquid lithium. NbN_0.9O_0.1 was observed on the surface of niobium specimens in test I and test II, and Nb₂N in test II and test III. The surface of the stainless-steel specimens in test II and test III was depleted with nickel and chromium elements, and deteriorated. The corrosion rates of the test specimens in test III were about 2, 5, 26 and 22 μm/yr for molybdenum or TZM, niobium, type 304 stainless steel and type 316, respectively. Nickel and Hastelloy N were severely attacked by liquid lithium at 600°C. These results were obtained for liquid lithium with a high nickel concentration.