A review of work related to reheat cracking in nuclear reactor pressure vessel steels

This review was completed in 1976 and describes the then-current knowledge on the problem of reheat cracking in weldments of nuclear pressure vessel steels. The incidence of underclad cracks and structural weld cracks in heat affected zones (HAZ) is described and current theories on the cracking mechanism and controlling factors discussed. Problems concerned with detecting the cracks by destructive and non-destructive methods are outlined. The paper then deals in detail with testing techniques for reheat crack susceptibility and with methods of controlling the problem of underclad and structural weld cracking. Finally, the engineering significance of the cracks in relation to hydrotesting and service is discussed. An appendix gives recommendations for additional work on the subject.     

A comparison of Western and Eastern nuclear reactor pressure vessel steels

The task was essentially to compare the irradiation response of `East' and `West' steels. Since the plates and forgings of pressure vessels must be welded together, it is obvious that the strength requirements of the welds and heat affected zones (HAZ) can be no less demanding than those of the plates and the forgings themselves, particularly as experience has shown that the most likely location for flaws is in the welds or their HAZs. These and the highly stressed regions of the reactor pressure vessel (RPV) are important because neutron irradiation degrades the mechanical properties of steels.After comparing the various designs, manufacture and materials of the various RPVs, a comparison was made of the irradiation response of these different steels. The role of mitigating the change in mechanical properties on irradiation by thermal annealing was also considered.Particular codes/guides could only be used for the predicting results underpinning their own database because a major difference between these national codes/guides is that the elements conferring irradiation sensitivity are different for the two cases considered, i.e. Russian codes [1] (PNAE G-7-002-86) and the USNRC guide [2] (RG 1.99 Rev. 2). In the former, copper and phosphorus are significant, while copper and nickel are identified as significant in the latter case.Predictions were compared for `real' materials used in NPPPVs whose compositions were known. The irradiation response of these steels is coincidentally similar. The essential difference in behaviour is in the lifetime fluence. Eastern steels are irradiated to a much higher fluence than Western steels. Differences in the predictions of the Eastern–Western codes/guides are a reflection of differences in the concentration of deleterious elements and pessimisms of the various codes/guides, particularly at low concentrations of deleterious elements where they are most conservative. Thirdly, and on a `fitness for purpose' basis, the shift in transition temperature produces a limitation to the lifetime of the earlier Eastern RPVs. However, by thermally annealing the RPV to mitigate the effect of neutron irradiation, where the conditions to recover the mechanical properties of both Eastern and Western steels are nearly the same, the operational life of these older Eastern plants has been extended. Life assurance of these plants has, therefore, become practicable.This aspect of RPV technology, which is currently being considered in the US, could extend the operational life of nuclear power plants and thereby reduce the cost of the electricity generated.     

An updated correlation for crack-arrest fracture toughness for nuclear reactor pressure vessel steels

An updated and statistically-rigorous correlation is provided for crack-arrest toughness values versus normalized temperature for light-water nuclear reactor pressure vessel (RPV) steels. The database used in this effort is larger than applied heretofore and includes results from tests of laboratory-size specimens and from tests of large-scale specimens, which contain features prototypical of operating RPVs. The mathematical methodology used is based on a lognormal distribution, with its parameters calculated by orthogonal distance regression. This correlation was developed as one of several items updated for use in the US Nuclear Regulatory Commission's extensive program to evaluate and potentially revise its rule for ensuring structural integrity of operating RPVs when subjected to pressurized thermal-shock transients.     

Analysis of biaxial loading effect on fracture toughness of reactor pressure vessel steels

The local stress–strain state (SSS) near the crack tip and its connection with the crack tip opening displacement and J-integral under biaxial loading have been studied by finite element methods in elastic–plastic finite strain statement. Numerical investigations have been performed for various crack lengths and two types of biaxial loading (tension and bending) under conditions of small- and large-scale yielding. To predict the biaxial loading effect on cleavage fracture toughness, the procedure has been elaborated, this being based on the revealed regularities for SSS near the crack tip under biaxial loading and brittle fracture criterion proposed earlier. Prediction of the biaxial loading effect on cleavage fracture toughness has been performed as applied to reactor pressure vessel steel. The calculated results have been compared with available experimental data. Alternative approaches for prediction of the biaxial loading effect on fracture toughness have been discussed.     

Application of advanced master curve approaches on WWER-440 reactor pressure vessel steels

The master curve (MC) approach used to measure the transition temperature, T₀, was standarised in the ASTM Standard Test Method E 1921 in 1997. The basic MC approach for analysis of fracture test results is intended for macroscopically homogeneous steels with a body centred cubic (ferritic) structure only. In reality, due to the manufacturing process, the steels in question are seldom fully macroscopically homogeneous.     

Modeling for ductile-to-brittle transition under ductile crack growth for reactor pressure vessel steels

The local stress–strain state (SSS) near the crack tip is investigated by the finite element method in the finite strain statement (with regard to a change of the crack tip blunting) for both stationary cracks and crack growing by a ductile mechanism. Using the revealed particularities of SSS near the stationary and growing crack tips and the local cleavage fracture criterion the phenomenon of the ductile-to-brittle transition is explained for reactor pressure vessel steels. The model is proposed to predict the amount of ductile crack extension preceding the ductile-to-brittle transition as a function of the test temperature. The procedure for calculation of the cleavage fracture toughness is also elaborated with regard to ductile crack extensions. Analysis of the obtained calculated results and available experimental data is made. Alternative approaches for the interpretation of the ductile-to-brittle transition are discussed.     

Fatigue crack growth behavior of reactor pressure vessel steels in air and high-temperature water environments

Fatigue tests under constant amplitude load were conducted on CT specimens of A533B3 steels with four levels of sulfur content at different temperatures in air and high-temperature water environments. A modified capacitance-type COD gauge was shown to be suitable for fatigue crack length measurement at high temperatures in air. The observation that the Young's moduli measured at a strain rate of 4 × 10⁻³ s⁻¹ for the A533B3 steels at 150 °C and 300 °C did not decrease with an increase in temperature seemed to be related to the presence of dynamic strain aging. The fatigue crack growth rates at 150 °C and 300 °C in air were about two and half times slower than those tested at 400 °C, because dynamic strain aging prevailed at 150 °C and 300 °C. Fractographic examination results suggested that inclusions embedded in secondary cracks enhanced the fatigue crack initiation rather than the fatigue crack growth. The fatigue crack growth rates taken in the oxygen-saturated water environment were one order of magnitude faster than those obtained in air.     

Recent Japanese research activities on probabilistic fracture mechanics for pressure vessel and piping of nuclear power plant

This paper describes a review of recent Japanese activities on probabilistic fracture mechanics (PFM) analyses. Japan Atomic Energy Agency (JAEA: previously JAERI) had sponsored research committees on PFM organized by Japan Society of Mechanical Engineers (JSME) and Japan Welding Engineering Society (JWES) for more than a decade. This work still continues with the same members in JWES. The purpose of the continuous activity is to provide probabilistic approaches in several fields of integrity problems of nuclear power plant. This paper shows some of the newest results of the JWES research committee. First topic is evaluation of the new JSME code case with rules of Fitness-For-Service from the view of PFM, including reactor pressure vessel subject to pressurized thermal shock loading, piping with a crack of the allowable size and effect of sizing accuracy for piping integrity. The next one is development of new PFM techniques including reliability assessment of piping with domestic (Japanese) SCC data and maintenance optimization of LWRs based on risk and economic models. The last topic is the international round robin program just starting from 2008.     

Stress corrosion cracking of power boiler drums

This paper deals with the study, analysis and technical diagnosis fundamentals concerning damage induced by stress corrosion cracking. The main repair and safe operation methods for power boiler drums are described; this work being based on plant experience.     

Radiation embrittlement modelling for reactor pressure vessel steels: I. Brittle fracture toughness prediction

Modelling for the irradiation effect on brittle fracture toughness of reactor pressure vessel (RPV) steel is performed on the basis of the probabilistic model for fracture toughness prediction proposed by the authors earlier. The irradiation effect on parameters controlling plastic deformation and brittle fracture of RPV steels is analyzed. The physical mechanisms are considered which control the cleavage microcrack nucleation for RPV steels in the unirradiated and irradiated states and also in state after post-irradiation annealing. Prediction of the temperature dependence of brittle fracture toughness is performed as applied to irradiated 2.5Cr–Mo–V reactor pressure vessel steel. Modelling of the fluence effect and the phosphorus and copper content effect on brittle fracture toughness is carried out. It is shown that the probabilistic model based on a new formulation for brittle fracture criterion allows the adequate modelling for the irradiation effect on fracture toughness for RPV steel. Application of alternative models is discussed for fracture toughness prediction for irradiated RPV steels.     

Radiation embrittlement modelling for reactor pressure vessel steels: II. Ductile fracture toughness prediction

Modelling for the irradiation effect on ductile fracture toughness of reactor pressure vessel steels (RPV) is performed on the basis of ductile fracture criterion proposed earlier by the authors. The irradiation effect on mechanisms controlling ductile fracture is considered from a physical viewpoint. Modelling of the irradiation effect is carried out on the critical strain for smooth cylindrical specimens and on the local critical strain for cracked specimens. On the basis of the performed studies a scheme that allows an evaluation of the upper shelf level of the K_IC(T) curve for irradiated RPV steels is proposed.     

Weibull statistical models of KIc/KIa fracture toughness databases for pressure vessel steels with an application to pressurized thermal shock assessments of nuclear reactor pressure vessels

This paper presents new statistical representations of recently extended fracture toughness K_Ic and K_Ia databases for pressure vessel steels. These models were developed by the Heavy Section Steel Technology program at Oak Ridge National Laboratory in support of the current effort by the U.S. Nuclear Regulatory Commission to update its regulatory guidance for pressurized-thermal-shock (PTS) transients in nuclear reactor pressure vessels. The Weibull distribution, with two of its parameters calculated by the Method of Moments point-estimation technique, forms the basis for the new statistical models. An application of the new K_Ic/K_Ia models, as implemented in the favor probabilistic fracture mechanics computer program, is also presented for three PTS transients.     

Hydrogen embrittlement of CrMo and CrMoV pressure vessel steels

The variation of $\text{K}{}_{\mathrm{i}}$ with time to fracture and the threshold values of $\text{K}{}_{\mathrm{ISH}}$ in an hydrogen environment were measured for four CrMo and CrMoV pressure vessel steels in various conditions of heat treatment. The CrMoV grades displayed higher $\text{K}{}_{\mathrm{ISH}}$ thresholds, suggesting that they are less susceptible to hydrogen embrittlement than CrMo grades of comparable strength. The findings were analysed with regard to the concurrent presence of various alloying elements and their effects on the microsegregation of detrimental impurity elements at the austenitic grain boundaries, where these harmful elements can, in conjunction with hydrogen, cause intercrystalline embrittlement. Studies were also made of the kinetics of stable crack growth in these materials.     

Mechanical methods of improving resistance to stress corrosion cracking in BWR piping systems

Pipelocks and the mechanical stress improvement process (MSIP) have been applied in BWR plants. Pipelocks restore the integrity of the weldments with identified cracks. MSIP removes residual tensile stresses from weldments, thus preventing initiation of cracks or retarding growth of pre-existing flaws in piping systems. The first 12-in pipelock installed at Commonwealth Edison's Quad Cities plant was inspected after operating for 18 months. Pipelocks for 10-in, 12-in and 28-in reactor safe-ends were fabricated for Carolina Power & Light's Brunswick plant. MSIP was applied at Commonwealth Edison's Dresden and LaSalle BWR plants. Extensive qualification has been completed for MSIP under US Nuclear Regulatory Commission and Electric Power Research Institute sponsorship. Sectioning of the pipe wall by Argonne National Laboratories provided stress distribution before and after MSIP for the 12-in and 28-in pipes. Measured ‘as welded’ tensile stresses before MSIP were within the range 30–50 000 psi. Compressive stresses after MSIP at the inside surface of the weldment reached more than 30 000 psi in both hoop and axial directions. The axial compressive stresses extended to the middle plane of the wall. Hoop stresses remained compressive through the wall. The stresses were uniformly distributed around the circumference of the pipe. J.A. Jones Applied Research Company completed an evaluation of MSIP applied to a precracked weldment between 28-in pipe and elbow. The pipe was squeezed to about 1·7% in the presence of cracks 25%, 50% and 90% through-wall. High compressive stresses were measured after MSIP. The cracks did not extend and could be identified after completion of the process by the usual UT technique. The use of mechanical methods becomes especially adequate for reactor safe-ends including bi- or tri-metallic joints. The use of overlay technique or induction heat stress improvement is more difficult due to high thermally induced strains at the strong discontinuity interface between materials of different thermal expansion. Basic concepts and practical application of mechanical methods to inhibit stress corrosion attack are described.     

Failure probability and optimum microstructure of pressure vessel steels

This paper presents a method for assessing the probability of brittle fracture in steels used for nuclear power generating equipment. The method is based on a statistical analysis of microstructural parameters and on physical models of the initiation and propagation of cleavage microcracks. The least probability of brittle failure ascertained by this method has been utilized for determining the optimum ferrite grain size. This method represents a foundation for systematic control of the brittle fracture characteristics of engineering steels employed for the components of nuclear power stations.     

Stress intensity factor solutions for arbitrarily shaped surface flaws in reactor pressure vessel nozzle corners

Extensive experimental results—based on the frozen stress photoelasticity technique for extracting stress intensities—for nozzle corner cracks in ITV and BWR geometries were reported by Smith et al.¹ Based on the above experimental studies, it was conjectured that if the crack shape inserted into a finite element model is not a real one, or if the inner fillet (for shallow flaws) or the outer boundary shape (for moderate to deep flaws) is improperly approximated, the obtained numerical results for stress intensity factors may differ significantly from the physical behaviour at the nozzle-vessel junction. On the other hand, almost all the numerical analyses published to date, based on finite elements, boundary integral equations or alternating techniques, considered only quarter-circular nozzle corner cracks.This paper presents stress intensity factor solutions for naturally shaped nozzle corner cracks in pressurised ITV and BWR vessels. Several actual crack geometries observed in the experimental work of Smith et al., cited above, are studied using the three-dimensional hybrid crack-element approach of Atluri et al.⁸ and Atluri and Kathiresan⁹ wherein the stress intensity factors and their variation along an arbitrarily shaped 3-D crack front are directly computed. In order to be able to compare the present results with the photoelastic experimental results (wherein the Poisson's ratio of the material is 0·5), some of the present numerical results are obtained for $\text{ν ? 0·5}$.In addition, some new solutions for stress intensity factors for pressurised thin (outer to inner radii ratios of $\text{～1·1}$) cylindrical vessels with belt-line flaws of semi-elliptical shapes of various aspect ratios and depth ratios are presented. Cases of surface flaws in the meridional direction—as well as in the circumferential direction—of the vessel are treated.     

Fracture toughness and other mechanical metallurgy characteristics of pressure vessel steels

The objective of this work is to investigate the effect of specimen size and evaluation method on estimates of fracture toughness for three commercial pressure vessel steels in different heat treatment conditions. It is found that the fracture toughness values K_C recorded on large, 150 mm thick, specimens were consistently close to the lower bound envelope of the K_δC, K_JC and K_E values established on 20 mm thick specimens. Comparison of the various elasto-plastic fracture criteria ascertained on a single specimen shows that the variations in these values are relatively slight which implies that the evaluation techniques used in this study can be considered as giving results that are, for practical purposes, the same.     

Combination of radiation and hydrogen damage of reactor pressure vessel materials

The hydrogen embrittlement of low-alloyed base steel, austenitic cladding and heat affected zone (HAZ) of a reactor pressure vessel was measured for both unirradiated and irradiated materials. The fracture toughness decreased with both hydrogen charging and neutron irradiation; the shift of the fracture toughness-temperature transient curve is influenced by both damage processes. The plastic zone in hydrogen-charged material becomes smaller. The total elongation of both CrMoV and CrNiMoV HAZ decreases with increasing hydrogen content. This influence is pronounced in the HAZ after a weld process without subsequent annealing, a total loss of plasticity being observed in this case. The properties of the austenitic layer are not influenced at comparable hydrogen contents.     

A new approach to estimate irradiation embrittlement of pressure vessel steels

The idea of an engineering version of the local approach to brittle fracture as well as the possibility for it to be employmed to estimate irradiation embrittlement of pressure vessel (PV) steels is considered. Unlike the conventional temperature-shift-based methodology, the approach presented utilises the concept of stability of the ductile state of the metal. A new characteristic “parameter of mechanical stability” P_ms is proposed. This characteristic enables quantification of the level of stability of the ductile state of an irradiated PV steel in a specimen or in a reactor vessel with a crack at the specified level of loading. Within the framework of the proposed concept, a value for end-of-life fluence for a reactor PV is predicted by the condition of exhaustion of stability of a ductile state of a steel ahead of the crack (P_ms=1).