Reactor Pressure Vessel Steels Research Articles

Verification of the master curve concept (ASTM E1921) and inhomogeneity analysis of a German RPV weld for various loading rates

Fracture-mechanical investigations were carried out on the weld metal of a German reactor pressure vessel steel 22NiMoCr3-7. The crack tip loading rate and test temperature were varied in the brittle-ductile transition region, while the results were evaluated using the Master Curve methodology (ASTME1921), including inhomogeneity analyses in Appendix X5. The highly inhomogeneous datasets could be uniformly described with a bimodal distribution function with good accuracy compared to the standard procedure. However, this agreement decreases with increasing crack tip loading rate. In contrast, a modified Master Curve with a shape factor of p = 0.03/°C progressively matches the data better with increasing crack tip loading rate. The weld metal has a significantly higher toughness than the base material. SEM analysis suggests that the inhomogeneity is due to the stochastic distribution of sharp microcracks, while the observed inhomogeneity is additionally superimposed by dynamic effects (adiabatic heating and local crack arrest) at increased crack tip loading rates. It is argued that the microstructural inhomogeneity is hereby »neutralized« and could instead be interpreted as a possible artifact of the standard Master Curve under dynamic conditions.

Research Progress of Steels for Nuclear Reactor Pressure Vessels.

The nuclear reactor pressure vessel is an important component of a nuclear power plant. It has been used in harsh environments such as high temperature, high pressure, neutron irradiation, thermal aging, corrosion and fatigue for a long time, which puts forward higher standards for the performance requirements for nuclear pressure vessel steel. Based on the characteristics of large size and wall thickness of the nuclear pressure vessel, combined with its performance requirements, this work studies the problems of forging technology, mechanical properties, irradiation damage, corrosion failure, thermal aging behavior and fatigue properties, and summarizes the research progress of nuclear pressure vessel materials. The influencing factors of microstructures evolution and mechanism of mechanical properties change of nuclear pressure vessel steel are analyzed in this work. The mechanical properties before and after irradiation are compared, and the influence mechanisms of irradiation hardening and embrittlement are also summarized. Although the stainless steel will be surfacing on the inner wall of nuclear pressure vessel to prevent corrosion, long-term operation may cause aging or deterioration of stainless steel, resulting in corrosion caused by the contact between the primary circuit water environment and the nuclear pressure vessel steel. Therefore, the corrosion behavior of nuclear pressure vessels materials is also summarized in detail. Meanwhile, the evolution mechanism of the microstructure of nuclear pressure vessel materials under thermal aging conditions is analyzed, and the mechanisms affecting the mechanical properties are also described. In addition, the influence mechanisms of internal and external factors on the fatigue properties, fatigue crack initiation and fatigue crack propagation of nuclear pressure vessel steel are analyzed in detail from different perspectives. Finally, the development direction and further research contents of nuclear pressure vessel materials are prospected in order to improve the service life and ensure safe service in harsh environment.

Fracture Toughness Evaluation of the Heat-Affected Zone Under the Weld Overlay Cladding in Reactor Pressure Vessel Steel

Abstract The fracture toughness in the heat-affected zone (HAZ), which is located under the weld overlay cladding on the inner surface of the reactor pressure vessel (RPV), was evaluated by considering the inhomogeneous microstructures of the HAZ. In this regard, 0.4T-C(T) specimens and miniature C(T) (Mini-C(T)) specimens were manufactured to clarify the effect of the HAZ on fracture toughness. First, the fracture toughness was evaluated using 0.4T-C(T) specimens targeted on microstructural distribution due to the cladding bead location. The reference temperature (To) of HAZ at middle region of cladding bead was 12 °C higher than that at overlap cladding beads. Second, the fracture toughness was evaluated using Mini-C(T) specimens focused on the microstructural difference with distance from the boundary between the cladding and the base metal. The To value of the central position located at 3–7 mm from the boundary between the cladding and the base metal was higher than those of the other positions in HAZ. However, the To values of HAZ for both 0.4T-C(T) and Mini-C(T) specimens were significantly lower than that of the base metal at a quarter thickness by 40–60 °C. Compared with the literature data that indicated fracture toughness at the inner surface without overlay cladding and the base metal at a quarter thickness, this study concluded that the HAZ under the overlay cladding showed no detrimental effect on the conservatism of the structural integrity assessment of RPV steel.

Investigation of Potential Material Inhomogeneity in the Magnetically Detected Neutron-Irradiation-Generated Structural Degradation of Nuclear Reactor Pressure Vessel Steel

A novel nondestructive method called magnetic adaptive testing has been previously applied to detect the neutron-irradiation-generated structural changes in reactor pressure vessel steel material. This method has been found to be a useful tool for this purpose, and good correlation—as a tendency—has been found between the estimated ductile-to-brittle transition temperature and magnetic parameters. However, a significant scattering of measured points was also observed for the investigated set of Charpy specimens. The main result of the work was that by magnetic selection of samples, the scatter can be notably reduced. As a conclusion, the magnetically measured parameters seemed to be precise and reliable for the detection of embrittlement of the reactor pressure vessel steel, with lower scattering of points than in the conventionally used destructive mechanical characteristics (ductile-to-brittle transition temperature). This result is surprising and needs further verification. The purpose of the present work is to repeat the measurement on irradiated reactor steel blocks. In this work, instead of the DBTT transition temperature, individually measured Vickers hardness (VH) data were used to help characterize the mechanical properties of the material. The so-called “property transformation” is a known and applied technique in the nuclear industry. The mechanical property characterized by the transition temperature cannot be determined individually for each specimen; instead, it can be obtained only on a set of samples by statistical fitting. Therefore, the individually measured Vickers hardness values can be utilized in order to predict the individual transition temperature values by the help of the property transformation technique. In this paper, however, not these derived transition temperature values, but their origins, the Vickers hardness values, are studied in a direct manner. The same behavior of blocks was observed as in the case of Charpy specimens, which is considered to validate the previously published results. As a possible reason for the scattering of points, large magnetic inhomogeneity of samples cut even from the same block was also proved. The magnetic parameters and Vickers hardness correlate well with each other. This result justifies the potential future application of magnetic techniques in practice aimed at the regular inspection of nuclear reactors.

Interplay between carbon and substitutional solute atoms during precipitation of Mn/Ni/Si-rich phases in model reactor pressure vessel steels

Trace C atoms can significantly alter the mechanical properties of steels, however, the effects of C atoms during precipitation of the Mn/Ni/Si-rich phases in reactor pressure vessel (RPV) steels remains unclear. Here, the interaction of C atoms with Mn/Ni/Si atoms during aging is investigated via the internal friction technique. Two Snoek peaks were detected: one is a superposition of the Snoek relaxation in Fe-C-Fe and the Snoek relaxation in Fe-C-Mn, and the other is the Snoek relaxation in Fe-C-Ni(Si). From the height variation of Snoek peaks with aging time, it was found that with the increasing aging time, the C atoms combine with the substitutional solute atoms at first, and then co-precipitate as MnSiC clusters at the GBs after 20 h aging at 600 °C. The variation of Snoek peak with aging time provides a reliable method to non-destructively detect the aggregation and precipitation of the M/Ni/Si-rich phases in RPV steels.

Identification and evolution of ultrafine precipitates in Fe-Cu alloys by first-principles modeling of positron annihilation

Understanding the formation and evolution of Cu precipitates in Fe-based alloys is crucial as they are key factors responsible for hardening and embrittlement. Dilute FeCu alloys are model materials for structural components in various application areas, including that of reactor pressure vessel steels for light water nuclear reactors. Positron annihilation spectroscopy (PAS) is a powerful tool to study the nucleation stage of both homogeneous and heterogeneous Cu precipitation, which are beyond the reach of most other experimental techniques. In this work, we present a first-principles study of positron annihilation in Fe-Cu systems. The positron annihilation characteristics (positron lifetimes, Doppler broadening spectra) are calculated for both homogeneous vacancy-free Cu clusters and heterogeneous vacancy-Cu complexes using two-component density functional theory. The theoretical results excellently agree with the available reference PAS experimental results. Our calculations show that the types of Cu precipitates can be clearly distinguished by positron annihilation, and the sizes of Cu precipitates can also be reasonably well estimated with our calculations. Moreover, we also successfully analyze the evolution of the experimental signals during isochronal annealing where the small Cu clusters change character. This work improves the understanding of the early-stage Cu precipitation in Fe matrix.

Contribution of dislocation loop to radiation-hardening of RPV steels studied by STEM/EDS with surveillance test pieces

Microstructure and radiation-hardening of surveillance test pieces of reactor pressure vessel steels have been studied by STEM/EDS and Vickers micro-hardness measurements. In a low-Cu steel (0.06 wt%), the microstructures irradiated with fluence of 5.62 × 1023 n/m2 were found to include nano-size Ni–Si-rich solute clusters (SCs) and dislocation loops (DLs). Most of the DLs were sustained after post-irradiation annealing at 500 °C for 0.5 h, while the number density of SCs decreased. The stability of the DLs is suggestive of entanglement with solute atoms. On the other hand, Cu-rich SCs were observed in high-Cu steels (0.17 wt%) irradiated with fluence of 1.01 × 1023 n/m2 and 6.11 × 1023 n/m2, however, no DLs observed. The Cu content in the matrix seems to be crucial for the nucleation of the DLs, which suggests that the mobility of the vacancy coupled with Cu atom dominates the nucleation/annihilation of self-interstitial atom. The contribution of DLs and SCs are discussed on the basis of the Russell-Brown and Orowan models, respectively.

A probabilistic approach for cleavage fracture including the statistics of microcracks: Application to a reactor pressure vessel steel

This work describes a probabilistic framework for cleavage fracture incorporating the measured statistics of microcracks to characterize the cleavage fracture toughness distribution in structural ferritic steels. Fracture toughness testing conducted on standard compact tension C(T) specimens for a 22NiMoCr37 pressure vessel steel, known as the EURO steel A, for varying test temperatures provides the cleavage fracture resistance data needed to determine the experimentally measured Jc-distribution. Metallographic examination of etched surfaces for the EURO steel also provides the distribution of carbides, which are assumed as the Griffith fracture-initiating particles, dispersed in the material from which the cleavage fracture toughness distribution is predicted. Overall, the analyses conducted in the present work show that the probabilistic model incorporating the statistics of microcracks holds significant promise as an engineering-level procedure to predict the fracture behavior in structural components with diverse range of temperature and possibly crack-tip constraint.

Numerical prediction of the effect of irradiation on the Charpy Upper Shelf Energy of Reactor Pressure Vessel steels

Neutron irradiation affects the mechanical properties of Reactor Pressure Vessel (RPV) steels used in Light Water Reactors (LWR). Fracture properties are particularly monitored based on surveillance programs, mostly through Charpy impact tests. The evolutions of Ductile Brittle Transition Temperature (DBTT) and Upper Shelf Energy (USE) with irradiation are evaluated and used for structural integrity assessment depending on regulation procedures. Semi-empirical correlations have been proposed to predict DBTT and USE shifts as a function of material chemical composition and fast neutron fluence, based on extended experimental database. In this study, numerical simulations of Charpy impact tests are performed in order to predict quantitatively the evolution of USE with irradiation. In a first part, the mechanical properties of an A508 Cl3 RPV steel are presented, including tensile tests on flat and notched samples and Charpy impact tests, for various irradiation levels and test temperatures. USE values are found to be within the bounds of correlation curves. In a second part, the experimental data are used to calibrate constitutive equations describing the material behavior up to high strain levels. Charpy impact test finite element simulations are then detailed, including ductile fracture modelling. In a third part, the ductile fracture model is calibrated based on the experimental data, and numerical simulations are used to predict the evolution of USE with irradiation. The decrease of USE with irradiation is recovered, and the numerical results are found to be in quantitative agreement with the correlation curves. Additional simulations confirm the agreement for different reference unirradiated USE and chemical composition, as well as using a simplified ductile fracture model. These results indicate that numerical simulations of Charpy impact test relying on calibrated constitutive equations can be used to predict USE and required experimental characterization are finally discussed.

Multiscale modelling in nuclear ferritic steels: From nano-sized defects to embrittlement

Radiation-induced embrittlement of nuclear steels is one of the main limiting factors for safe long-term operation of nuclear power plants. In support of accurate and safe reactor pressure vessel (RPV) lifetime assessments, we developed a physics-based model that predicts RPV steel hardening and subsequent embrittlement as a consequence of the formation of nano-sized clusters of minor alloying elements. This model is shown to provide reliable assessments of embrittlement for a very wide range of materials, with higher accuracy than industrial correlations. The core of our model is a multiscale modelling tool that predicts the kinetics of solute clustering, given the steel chemical composition and its irradiation conditions. It is based on the observation that the formation of solute clusters ensues from atomic transport driven by radiation-induced mechanisms, differently from classical nucleation-and-growth theories. We then show that the predicted information about solute clustering can be translated into a reliable estimate for radiation-induced embrittlement, via standard hardening laws based on the dispersed barrier model. We demonstrate the validity of our approach by applying it to hundreds of nuclear reactors vessels from all over the world.

Cluster dynamics modeling of Mn-Ni-Si precipitates coupled with radiation-induced segregation in low-Cu reactor pressure vessel steels

Formation of precipitates enhanced or induced by irradiation causes hardening and embrittlement of nuclear structural materials. Post-irradiation microstructure characterization of reactor pressure vessel (RPV) steels has shown that precipitation can be strongly associated with radiation-induced segregation (RIS) of solutes on dislocations. However, the effect of RIS on the kinetics of Mn-Ni-Si precipitation has not been quantified and coupled in previous model predictions. In this study, a new hybrid and spatially-dependent precipitation model is developed that couples cluster dynamics with RIS with transport coefficients of diffusion mediated by vacancies and interstitials. The new approach provides a unique way to account for concurrent evolution of heterogeneous cluster densities as well as solute and point defect concentration profiles. The model is applied to study the segregation of Mn, Ni, and Si on dislocations and heterogeneous nucleation of Mn-Ni-Si rich precipitates (MNSPs) in a low-Cu RPV steel. The RIS result shows that the segregation tendency is dominated by the vacancy mechanism for Mn, Ni, and Si. The coupled cluster dynamics modeling result shows that the onset of MNSP nucleation on dislocations depends strongly on RIS and occurs at the fluence of 2×1023n..m−2. The number density and mean radius can reach ∼1024m−3 and 1−2 nm at high fluence, respectively. These observations suggest that the role of RIS and heterogeneous nucleation can be significant for RPV steels under high neutron fluence. Results of simulations adopting various dose rates and dislocation densities show an increased fraction of MNSPs on dislocations at the condition of low dose rate, high dislocation density, and high fluence.

A study of isothermal mechanical fatigue of welded reactor steel (15H2MFA)

As far as the reactor pressure vessel steel is concerned, the isothermal fatigue study of welded joints is helpful to protect the pressure vessel from blasting and save the life of its operators. Furthermore, the study helps to improve the future design of the reactor components and also enthuse to develop new materials. In material science, the fatigue of metal is quite difficult to deal with, and that of welded joint is an even more complex phenomenon. This complexity arises from different main factors: thermal cycles during welding, which will strongly affect the base material and induce residual stresses and distortion, the fusion process with the filler metal leads to heterogeneity to microstructure which will affect or change the original mechanical properties and chemical composition through out of the welded joint. In this study, isothermal mechanical fatigue of welded reactor steel has been contemplated. The samples of the welded joint used in this study are the ‘15H2MFA steel type of VVER-440’ reactor pressure vessel operating at Paks Nuclear Power Plant, Hungary. The reactor vessel is made of low-alloy heat-resistant steel and austenitic steel with a corrosion-resistant inner (clad) surface.

Effectiveness of Thermal Annealing in Recovery of Tensile Properties of Compositionally Tailored PWR Model Steels Irradiated in LYRA-10

Understanding the mechanical behaviour of reactor pressure vessel (RPV) steels at high fluences has become an important topic in regard to Long-Term Operations (LTO) of existing nuclear power plants (NPP). The effectiveness of thermal annealing treatments to recover the mechanical properties of compositionally tailored pressurised water reactor (PWR) model steels irradiated to high neutron fluences, up to 1.22 × 1020 n·cm−2, is analysed in this study. Tensile testing of four different PWR RPV steels was performed after irradiation and subsequent recovery annealing treatment at 450 °C for 40 h. Irradiation-induced hardening and the effectiveness of recovery thermal annealing have been assessed by comparing the strength and ductility properties of irradiated and irradiated and subsequently annealed samples with unirradiated reference samples for all four model steel. The annealing treatment resulted in a significant recovery of the yield strength (~75–89%) and the ultimate tensile strength (~78–96%) of all four PWR model steels. This study proves that substantial irradiation-induced hardening (up to ~389 MPa) observed in steels containing high Ni and Mn contents can still be recovered using the thermal annealing treatment. No influence of annealing on ductility properties has been observed for all four model steels. Microscopy analyses of these steels to understand the underlying irradiation damage and recovery mechanisms are planned for the near future.

Multicriteria Analytical Model for Mechanical Integrity Prognostics of Reactor Pressure Vessels Manufactured from Forged and Rolled Steels

The aim of this work is to present a new analytical model to evaluate jointly the mechanical integrity and the fitness-for-service of nuclear reactor pressure-vessels steels. This new methodology integrates a robust and regulated irradiation embrittlement prediction model such as the ASTM E-900 with the ASME Fitness-for-Service code used widely in other demanding industries, such as oil and gas, to evaluate, among others, the risk of experiencing degradation mechanisms such as the brittle fracture (generated, in this case, due to the irradiation embrittlement). This multicriteria analytical model, which is based on a new formulation of the brittle fracture criterion, allows an adequate prediction of the irradiation effect on the fracture toughness of reactor pressure-vessel steels, letting us jointly evaluate the mechanical integrity and the fitness-for-service of the vessel by using standardized limit conditions. This allows making decisions during the design, manufacturing and in-service of reactor pressure vessels. The results obtained by the application of the methodology are coherent with several historical experimental works.

Capturing the Temperature Dependence of Cleavage Fracture Toughness in the Ductile-to-Brittle Transition Regime in Ferritic Steels Using an Improved Engineering Local Approach

Abstract Ferritic steels, which are typically used for critical reactor components, including reactor pressure vessels (RPV), exhibit a temperature-dependent probability of cleavage fracture, termed ductile-to-brittle transition. The fracture process has been linked to the interaction between matrix plasticity and second-phase particles. Under high-enough loads, a competition exists between cleavage and ductile fracture, which results from particles rupturing to form microcracks or particles decohering to form microvoids, respectively. Currently, there is no sufficiently adequate model that can predict accurately the reduced probability of cleavage with increasing temperature and the associated increase of plastic deformation. In this work, failure probability has been estimated using a local approach to cleavage fracture incorporating the statistics of microcracks. It is shown that changes in the deformation material properties are not enough to capture the significant changes in fracture toughness. Instead, a correction to the fraction of particles converted to eligible for cleavage microcracks, with an exponential dependence on the plastic strains, is proposed. The proposed method is compared with previous corrections that incorporate the plastic strains, and its advantages are demonstrated. The method is developed for the RPV steel 22NiMoCr37 and using experimental data for a standard compact tension C(T) specimen. The proposed approach offers more accurate calculations of cleavage fracture toughness in the ductile-to-brittle transition regime using only a decoupled model, which is attractive for engineering practice.

Machine learning predictions of irradiation embrittlement in reactor pressure vessel steels

Irradiation increases the yield stress and embrittles light water reactor (LWR) pressure vessel steels. In this study, we demonstrate some of the potential benefits and risks of using machine learning models to predict irradiation hardening extrapolated to low flux, high fluence, extended life conditions. The machine learning training data included the Irradiation Variable for lower flux irradiations up to an intermediate fluence, plus the Belgian Reactor 2 and Advanced Test Reactor 1 for very high flux irradiations, up to very high fluence. Notably, the machine learning model predictions for the high fluence, intermediate flux Advanced Test Reactor 2 irradiations are superior to extrapolations of existing hardening models. The successful extrapolations showed that machine learning models are capable of capturing key intermediate flux effects at high fluence. Similar approaches, applied to expanded databases, could be used to predict hardening in LWRs under life-extension conditions.

Remarkable embrittlement and its origins in quenched reactor pressure vessel steels

Reactor pressure vessels (RPVs) are one of the main barriers against nuclear accidents, therefore good toughness is indispensable for keeping the integrity of RPVs to prevent the leakage of radioactive substances. It is well known that high temperature and large thermal gradient of nuclear accidents would lead to embrittlement or even rupture of RPVs. However, how damages originate in severe nuclear accidents is still indistinct. This study is aimed to reveal the embrittlement mechanism under a large temperature gradient in quenched RPV steels. It is found that certain strengthening as well as remarkable embrittlement occurs after quenched, further investigations find it probably derive from the generation of plenty of Moiré fringes and dislocation networks, which are supposed to be generated by shearing of crystal planes. This paper reports the deterioration phenomenon caused by a large thermal gradient, reveals the origins remarkable embrittlement and indicates that extra attention must be aroused to take this thermal-stress-induced embrittlement into consideration in nuclear accidents analysis.

Evaluating the irradiation hardening of reactor pressure vessel steels by nanoindentation hardness test and micropillar compression test

ABSTRACT Ultra-small testing technologies (USTTs), such as nanoindentation hardness tests and micropillar compression tests, can be potentially used to measure the mechanical properties of irradiated materials. However, the hardness and yield strengths of USTTs are different from those of standardized methods on bulk samples. In addition, irradiation-induced defects can affect the size dependence on the mechanical properties. This study examined A533B reactor pressure vessel steels before and after 5 MeV electron irradiation using USTTs. The depth dependence of the nanoindentation hardness was analyzed using the Nix-Gao model to evaluate the bulk-equivalent nanoindentation hardness and confirm its good linear relationship with the Vickers hardness. The compression yield strength was evaluated with a truncated square pyramid shape micro-pillars where the edge length (d) of the top surface and the height (L) of the micropillars were made similar: (d, L) = (1,2), (2,4), (3,6), (4,8), and (8,16) in μm. The size dependence of the yield stress in the micropillars has a minimum at d = 2 µm, which remained unaffected by electron beam irradiation up to 22 mdpa. The average yield stress of the d = 4 and 8 µm specimens was similar to that of the bulk tensile specimens.

Machine-Learning Approach to Determine Surface Quality on a Reactor Pressure Vessel (RPV) Steel

Surface quality measures such as roughness, and especially its uncertain character, affect most magnetic non-destructive testing methods and limits their performance in terms of an achievable signal-to-noise ratio and reliability. This paper is primarily focused on an experimental study targeting nuclear reactor materials manufactured from the milling process with various machining parameters to produce varying surface quality conditions to mimic the varying material surface qualities of in-field conditions. From energising a local area electromagnetically, a receiver coil is used to obtain the emitted Barkhausen noise, from which the condition of the material surface can be inspected. Investigations were carried out with the support of machine-learning algorithms, such as Neural Networks (NN) and Classification and Regression Trees (CART), to identify the differences in surface quality. Another challenge often faced is undertaking an analysis with limited experimental data. Other non-destructive methods such as Magnetic Adaptive Testing (MAT) were used to provide data imputation for missing data using other intelligent algorithms. For data reinforcement, data augmentation was used. With more data the problem of ‘the curse of data dimensionality’ is addressed. It demonstrated how both data imputation and augmentation can improve measurement datasets.

Relation of embrittlement to phosphorus grain-boundary segregation for an advanced Ni–Cr–Mo RPV steel

Embrittlement is an essential issue for reactor pressure vessel (RPV) steels. Grain-boundary segregation (GBS) of impurity phosphorous is one of the major reasons for non-hardening embrittlement of an RPV steel. The main aim of the present work is to establish a relation of embrittlement to phosphorous GBS for an advanced SA508-4N Ni–Cr–Mo RPV steel so that one can forecast its embrittling tendency. To obtain this type of relation, the steel samples doped with phosphorous are quenched from 890 to 1100 °C to obtain different prior austenite grain sizes (PAGSs) and then tempered at 650 °C, followed by ageing at different lower temperatures so as to achieve different phosphorous segregation levels at prior austenite grain boundaries. The characterization work is done with the use of mini-Charpy impact tests (specimen: 2.5 mm × 2.5 mm in cross section) together with Auger electron spectroscopy, scanning electron microscopy, and optical microscopy. The results indicate that there is a linear relationship between ductile-to-brittle transition temperature (DBTT/oC) and phosphorous GBS concentration (Cp/at.%) for both cases (quenching from 890 °C: PAGS = 34 μm; quenching from 1100 °C: PAGS = 112 μm). The relations can be expressed as DBTT=13.13Cp−335.70 (PAGS = 34 μm) and DBTT=6.69Cp−223.87 (PAGS = 112 μm). The rate of material embrittlement, characterized by the slope of the line, is associated with both the phosphorous segregation-induced grain-boundary (GB) embrittlement and the total GB area per unit volume. Interestingly, there is a critical phosphorous GBS concentration, below which the DBTT of the coarse-grained specimen is higher due to its higher intrinsic brittleness and above which that of the fine-grained specimen is higher at the same phosphorous GBS level due to the considerable phosphorous segregation-induced GB embrittlement along with a much larger total GB area per unit volume.