Structural Integrity Assessment of a Nuclear Vessel through ASME and Master Curve Approaches Using Irradiation Embrittlement Predictions

The assessment of the structural integrity of nuclear vessels is based on a series of procedures developed in the 1970s and 1980s. On one hand, curves that, according to the American Society of Mechanical Engineers code, describe the relationship between steel toughness and temperature in the ductile-to-brittle transition region, based on the reference temperature concept RTNDT, were adopted in 1972. On the other hand, the material embrittlement derived from the exposure of steel to neutron irradiation is determined through the model included in “Regulatory Guide 1.99 Rev. 2,” published in 1988. Since then, there have been enormous advances in this field. For example, the Master Curve, based on the reference temperature T0, describes the relationship between toughness and temperature in the transition zone more realistically and with much more robust microstructural and mechanical foundations and uses the elastic-plastic fracture toughness KJc. Moreover, improved models have been developed to estimate the embrittlement of steel subjected to neutron irradiation, such as ASTM E900, Standard Guide for Predicting Radiation-Induced Transition Temperature Shift in Reactor Vessel Materials. This study is aimed at comparing the results obtained using traditional procedures to the improved alternatives developed later. For this purpose, the behavior of the steel of a nuclear vessel that is currently under construction has been experimentally characterized through RTNDT and T0 parameters. In addition, the material embrittlement has been quantified using “Regulatory Guide 1.99 Rev. 2” and ASTM E900. These experimental results have been transferred to the assessment of the structural integrity of the vessel to determine Manuscript received August 7, 2018; accepted for publication September 27, 2018; published online January 8, 2019. 1 Laboratory of Science and Engineering of Materials, University of Cantabria, E.T.S. de Ingenieros de Caminos, Canales y Puertos, Av. Los Castros 44, Santander 39005, Spain (Corresponding author), e-mail: ferrenod@unican.es 2 Technological Centre for Components, Parque Científico y Tecnológico de Cantabria C/ Isabel Torres n° 1, Santander 39011, Spain 3 Laboratory of Science and Engineering of Materials, University of Cantabria, E.T.S. de Ingenieros de Caminos, Canales y Puertos, Av. Los Castros 44, Santander 39005, Spain 4 Equipos Nucleares, S.A., S.M.E. Av. Juan Carlos I, 8, Maliaño 39600, Spain Journal of Testing and Evaluation Copyright © 2019 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959 doi:10.1520/JTE20180569 available online at www.astm.org Copyright by ASTM Int'l (all rights reserved); Fri Sep 18 04:49:27 EDT 2020 Downloaded/printed by Universidad De Cantabria Biblioteca (Universidad De Cantabria Biblioteca) pursuant to License Agreement. No further reproductions authorized. the pressure-temperature limit curves and size of the maximum admissible defect as a function of the operation time of the plant. The results have allowed the implicit overconservatism present in the traditional procedures to


Introduction
Neutron irradiation is the most relevant source of degradation for nuclear reactor pressure vessel (RPV) steels.Because of the irradiation, the material fracture toughness decreases over time, leading to a shift in the ductileto-brittle transition (DBT) temperature.The surveillance program of the plant makes it possible to monitor changes in the fracture toughness of the vessel steel; from this information, the conditions under which the vessel can be operated throughout its in-service life can be determined.A surveillance program consists of placing capsules holding specimens fabricated with the same steel as that of the vessel in the beltline region (that is, the general area of the reactor vessel near the core midplane where radiation dose rates are at their highest) and attaching them to the inside wall of the vessel.Thus, the specimen irradiation history duplicates the neutron spectrum, temperature history, and maximum neutron fluence undergone by the reactor vessel's inner surface.
The fracture toughness requirements for ferritic materials of nuclear power plants (NPPs) must fulfill the acceptance and performance criteria of Appendix G of Section III of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code [1,2].According to this procedure, the fracture resistance of the vessel material in the DBT region before irradiation is described by the reference temperature RT NDT(U) , which is obtained from Charpy impact and Pellini drop weight tests through empirical and conservative correlations.The Master Curve (MC) is an alternative method based on direct characterization of the fracture toughness in the DBT region by means of K Jc tests.This approach is a consequence of the developments in elastic-plastic fracture mechanics (EPFM) together with an increased understanding of the micromechanisms of cleavage fracture.The basic MC method for analysis of brittle fracture test results is defined in ASTM E1921-17a, Standard Test Method for Determination of Reference Temperature, T 0 , for Ferritic Steels in the Transition Range (Superseded) [3], where the reference temperature T 0 is employed.T 0 is defined as the temperature at which the median fracture toughness obtained with B = 25.4 mm thickness (1T, according to ASTM terminology) specimens is 100 MPa • m 1/2 .The reference temperature T 0 completely characterizes the fracture toughness in the DBT region of ferritic steels that experience an onset of cleavage cracking at elastic or elastic-plastic K Jc instabilities.This fact is well supported by experience.
The predictions of these approaches (ASME and MC) have been compared in this study for the vessel steel of a boiling water reactor that is currently under construction.Thus, to quantify the level of inherent conservatism, two approaches have been compared in this work.First, Pressure-Temperature limit curves (P-T curves) have been obtained following the ASME code that consistently incorporates the MC approach.These curves relate the maximum allowable pressure with the operation temperature in order to avoid any risk of brittle fracture of the vessel for a postulated flaw with a depth equal to one quarter of the thickness of the vessel (t/4).Second, the allowable crack size has been calculated for the heat-up and cooldown operations as well as the hydrostatic test of the vessel.In this latter case, the stress intensity factors have been obtained by means of ASME Code Section XI [2] as well as the European Fitness-for-Service Network (FITNET) Fitness-for-Service (FFS) procedure [4].To achieve these goals, the fracture properties of the unirradiated base metal of the vessel in the DBT region were experimentally obtained.In this sense, Charpy and Pellini tests were carried out to apply the ASME code, whereas K Jc fracture toughness tests were performed to apply the FITNET FFS procedure.Furthermore, the fracture properties of the irradiated material were predicted by using a series of analytical models available in standards and literature.Then, an assessment of structural integrity was also carried out for the irradiated condition based on these models for neutron embrittlement.

REGULATORY METHOD REGARDING FRACTURE TOUGHNESS IN THE DBT REGION
Title 10 of the Code of Federal Regulations, 10CFR50 [5], imposes P-T limits on the reactor coolant pressure boundary for NPPs designed in the USA.According to this law, fracture toughness of ferritic materials must fulfill the acceptance and performance criteria of Appendix G of Section III of the ASME Boiler and Pressure Vessel Code [1].The toughness of the vessel steel in the DBT region before irradiation is described by means of the reference temperature RT NDT(U) , which is obtained through the combination of Charpy impact and Pellini drop weight tests.This parameter is used to index two generic curves [1,2] relating toughness versus temperature (see Eqs 1 and 2, in which temperatures must be expressed in °C and toughness is obtained in MPa • m 1/2 ).The K Ic curve describes the lower envelope to a large set of K Ic data, while the K IR is a lower envelope to a combined set of K Ic , K Id (dynamical tests), and K Ia (crack arrest tests) data, therefore being more conservative than the former.This approach is based on Linear Elastic Fracture Mechanics (LEFM) and assumes that the ASME curves are representative of a wide variety of vessel steels; large scatter, typical in the DBT region, is avoided by taking lower envelopes into account.Consequently, the method provides a deterministic and excessively conservative response in most cases.

THE MC
The MC approach was originally proposed by Wallin [6][7][8][9] to statistically describe the fracture toughness in the DBT region and is based on the experimental characterization of the fracture toughness.This method is a consequence of the developments in EPFM together with an increased understanding of the micromechanisms involved in cleavage fracture.The basic MC method for analysis of brittle fracture test results is defined in ASTM E1921-17a [3].The mathematical and empirical details of the procedure are available in Ref. [10].Some examples of successful applications can be found in Refs.[11][12][13][14][15][16].The main features and advantages of the method are hereafter summarized: • Fracture is governed by the weakest link statistics, which follows a three-parameter Weibull distribution.
One of the main advantages of the method is that it allows data from different-sized specimens to be compared.As thickness increases, the material toughness K Jc is reduced, which is due to the higher probability of finding a critical particle for the applied load.ASTM E1921-17a [3] provides Eq 3 to relate the fracture toughness, K Jc , for specimens of different thickness.
• Eq 4 provides the value of K Jc for a given cumulative failure probability, P f , once T 0 , the so-called MC reference temperature, has been determined.T 0 corresponds to the temperature such that the median fracture toughness for a 25.4-mm-thick specimen (1T) has the value 100 MPa • m 1/2 .The confidence bounds of the distribution (usually taking P f = 0.01 or 0.05 for the lower bound and 0.95 or 0.99 for the upper bound) can be obtained by applying Eq 4. As a particular case, the expression for the median fracture toughness (P f = 0.5) is determined as seen in Eq 5.
• A fitting procedure must be applied to find the optimum value of T 0 for a particular set of experimental data (K Jc and test temperature).For this task, all data are thickness-adjusted to the reference specimen thickness B 0 = 25.4 mm (1T) using Eq 3. The procedure can be applied either to a single test temperature or transition curve data.In the latter approach, T 0 is estimated from the size-adjusted K JC data (K JC,1T ) using a multitemperature maximum likelihood method.• This statistical analysis to estimate T 0 can be reliably performed even with a small number of fracture toughness tests (usually between six and ten specimens).Moreover, as an EPFM approach is used, the specimen size requirements are much less demanding than those of LEFM [17].These remarks are of great relevance for nuclear reactor surveillance programs for which the amount of material available is usually very limited and consists of small-size samples (Charpy V-notch (CVN) specimens).• To estimate the reference temperature T 0 , a previous censoring of the non-size-adjusted data must be applied.Fracture toughness data that are larger than the validity limit given by Eq 6, as defined in Ref. [3], are reduced to the validity limit, K Jc(lim) , and treated as censored values in the subsequent estimation stage.This condition is imposed to guarantee high constraint conditions in the crack front during the fracture process.
where s y is the yield stress at the test temperature, E is the Young's modulus, b 0 is the initial ligament, and ν is Poisson's ratio (ν = 0.3 in this case).It must be stressed that the factor of 30 in Eq 6 is currently under discussion [10] and that, for instance, ASTM E1820-17, Standard Test Method for Measurement of Fracture Toughness (Superseded) [18], imposes a more demanding limit with a factor of 50 or 100, depending on the nature of the steel.
• The standard deviation in the estimate of T 0 , expressed in °C, is given by Eq 7: where r represents the total number of valid specimens (uncensored results) used to establish T 0 .The values of factor β are provided in Ref. [3].
• As stated in Ref. [3], the reference temperature T 0 must be determined from quasistatic strain rates.
However, the experimental evidence shows that the MC concept can also be applied to dynamic tests [13,14,[19][20][21].In this sense, Wallin [19] has proposed a simple semiempirical expression for the strain rate dependence of T 0 .The error of the expression is only about ±20 %, covering yield strengths from 200 to 1,000 MPa.

CORRELATIONS TO PREDICT THE MATERIAL EMBRITTLEMENT
The decrease in material toughness that is due to neutron irradiation in the DBT region is currently estimated through empirical methods based on the shift undergone by Charpy impact curves that are obtained from the surveillance capsule specimens, which are retrieved periodically according to the plant withdrawal schedule.As stated in 10CFR50 [5], the effect of neutron fluence on the behavior of the material is predicted by "Regulatory Guide 1.99 Rev. 2" [22], which provides Eq 8 for the evolution of RT NDT : where ΔRT NDT represents the shift in the reference temperature that is due to irradiation (which is assumed to be equal to the shift of the Charpy transition curve indexed at 41 J; thus, ΔRT NDT = ΔT 41 J ).The third term, M, is the margin that must be added to obtain a conservative estimation.The procedure in Ref. [22] enables ΔRT NDT to be obtained even when no credible surveillance data are available by means of an equation based on the chemistry of the steel (copper and nickel content) and the neutron fluence received.
Moreover, in 2002, the ASTM committee approved ASTM E900-15e1, Standard Guide for Predicting Radiation-Induced Transition Temperature Shift in Reactor Vessel Materials [23], to predict the radiation-induced transition temperature shift (TTS) in reactor vessel materials.The embrittlement correlation proposed, which allows ΔT 41J to be estimated, is based on the copper and nickel content, irradiation temperature, and neutron fluence.The expression is mechanistically guided based on the current understanding of two mechanisms of embrittlement, stable matrix damage (SMD) and copper-rich precipitation (CRP).Saturation of copper effects (for different weld materials) was included.The mean value of the TTS is based on adding these two contributions, as stated in Eq 9: Several important features of the model are summarized as follows: • The SMD contribution depends on the irradiation temperature and neutron fluence.
• The CRP term is dependent on the copper and nickel content and neutron fluence, including a specific material factor (which distinguishes between weld, forgings, etc.).When the copper content is less than 0.072 wt.%, then CRP = 0, and the maximum copper content considered is 0.305 wt.% for welds.• The standard error of the correlation is 22.0°F (12.2°C).
• The neutron fluence rate is not included in the model.Although the surveillance capsule database includes neutron fluence rates ranging from 2 • 10 8 n/cm 2 /s to 1 • 10 12 n/cm 2 /s, the preponderance of the data lies between 3 • 10 9 n/cm 2 /s and 2 • 10 11 n/cm 2 /s.Within the limitations of the surveillance capsule database, a neutron fluence rate effect could not be unambiguously identified.
As a final remark, it is generally accepted that an approximate 1:1 correlation (slightly different for base and weld metals) does exist between the irradiation-induced shifts ΔRT NDT and ΔT 0 , as illustrated by an international database compiled by Sokolov and Nanstad from various publications [24].In this sense, the linear fitting ΔT 0 = 1.04 • ΔT 41 J is proposed (for a database of 126 data points, including both base and weld materials) with an uncertainty of 34°C.

AVAILABLE MATERIAL
The specimens and samples tested in this study were obtained from the base material of the surveillance semiring of an actual nuclear vessel that was recently manufactured (the plant is currently under construction).The surveillance ring was extracted from one of the forged rings that were subsequently welded to manufacture the beltline of the RPV.The base material is carbon steel grade SA-508M Grade 3 Class 1 by ASME Section II [25].This steel was subjected to normalizing-tempering plus quenching-tempering heat treatment after forging.The specimens employed in this study were selected to determine all the material parameters needed to conduct the structural integrity assessments (ASME versus MC).Specifically, the following tests were carried out: chemical composition tests, tension tests (as a function of temperature), Charpy impact tests (as a function of temperature), Pellini Drop-Weight (DW) tests, and K Jc fracture toughness tests (to determine the reference temperature T 0 ); moreover, the fracture surfaces were examined by means of Scanning Electron Microscopy (SEM).These experimental methods are described in the sections "Available Material" through "Fracture Toughness Tests" in "Materials and Methods."Following the recommendation of Ref. [26], the samples were obtained at depths of 1/4 and 3/4 of the thickness of the ring.Fig. 1 shows the distribution of specimens in the semiring; note that tension specimens were obtained in L and T orientations and Charpy specimens were obtained in LT and TL orientations.Table 1 summarizes the number and purpose of the specimens employed in this study.

TENSION TESTS
Tensile properties are necessary to evaluate the maximum measurement capacity in K JC fracture toughness tests, as seen in ASTM E1921-17a [3]; see Eq 7.Ten tension tests were carried out for each of the orientations at temperatures ranging from −145°C to 300°C by means of an Instron 8501 universal testing machine (Norwood, MA) at a fixed strain rate of 10 −3 min −1 .The tension tests were carried out according to ASTM E8 / E8M-16a, Standard Test Methods for Tension Testing of Metallic Materials [27].The load-elongation curves were recorded up to fracture by using extensometers designed to operate at low and high temperatures.

CHARPY IMPACT TESTS
The CVN impact tests were performed and analyzed according to ASTM E23-16b, Standard Test Methods for Notched Bar Impact Testing of Metallic Materials [28].Tests were carried out by means of an AMSLER RKP 300 (Zamudio, Spain) device with a loading capacity of 300 J.

PELLINI DW TESTS
The DW tests were performed following the requirements of ASTM E208-06(2012), Standard Test Method for Conducting Drop-Weight Test to Determine Nil-Ductility Transition Temperature of Ferritic Steels (Superseded) [29].The test machine, which was designed and manufactured in agreement with Ref. [29], is shown in Fig. 2.   The loading method consisted of applying a controlled rate of crack opening displacement as measured at the load line (0.06 mm/min ∼ 0.38 MPa • m 1/2 • s −1 ).The J-integral was calculated from the load-crack opening displacement curve at crack instability, in accordance with the ASTM standard [3].

SEM FRACTOGRAPHIC STUDY
The fracture surfaces of the K Jc tests were examined by SEM to ascertain the existence of cleavage as the physical fracture mechanism and to localize the initiation sites.The SEM device EVO MA 15 (Zeiss, Oberkochen, Germany) was employed.

CHEMICAL COMPOSITION
Table 2 shows the chemical composition of the vessel as well as the ranges and limits imposed by the specifications [25].A total of four chemical analyses were performed; then, the mean μ and standard deviation σ for each of the elements were obtained, as shown in Table 2.The main goal of these analyses is to determine the amount of those chemical elements that enhances the irradiation embrittlement, such as copper and nickel, in order to obtain the shift in DBT according to the models introduced in the "Correlations to Predict the Material Embrittlement" section.Note that the limits imposed by Section II of the ASME Code [25] for SA-508M Grade 3 Class 1 are satisfied in all cases.
TENSION TESTS Fig. 3 shows the experimental results, yield stress, and tensile strength of the tension tests carried out in L (Fig. 3a) and T (Fig. 3b) orientations.The temperature dependence of the yield stress s Y and ultimate strength s U were fitted using suitable expressions of the form s = C 0 + C 1 • exp(-C 2 • T) (where C 0 , C 1 , and C 2 are the fitting parameters).The results of the fittings are compiled in Table 3.
The data fittings are included in the figure.Knowledge of the tensile properties is necessary to evaluate the maximum measurement capacity, as seen in ASTM E1921-17a [3]; see Eq 8.As can be observed, the material response is highly isotropic since there are no significant differences between the L and T orientations.

CHARPY IMPACT TESTS
The results of the Charpy tests are represented in Fig. 4. The energy absorbed, E, is shown in Fig. 4a, in which the dashed lines represent relevant energy levels (27 J, 41 J, and 68 J) to define the DBT (e. g, T 28 J , T 41 J , and T 68 J ,    respectively).In Fig. 4b, the lateral expansion, LE, is represented against temperature; the dashed line corresponds to LE = 0.9 mm.Finally, in Fig. 4c, the shear fracture appearance, SFA, is plotted against temperature; SFA = 50 % is represented by means of a horizontal line.The experimental data were fitted by means of hyperbolic tangent curves of the form E (or LE) . These fittings were used to obtain a number of relevant results for LT and TL orientations that are summarized in Table 4, including the values of upper shelf energy (USE), lower shelf energy (LSE) as well as some representative transition temperatures (such as T 27 J and T 41 J , among others).
As can be seen, the fracture behavior in LT and TL orientations is very similar.The only discrepancy is observed in T 50 % (Table 4).However, this result should be interpreted as an artifact derived from the fitting of the data, as can be clearly appreciated in Fig. 4c.

PELLINI DW TESTS
The Pellini DW test provides the Nil-Ductility Temperature (NDT).The experimental results obtained in this study are shown in Table 5. RT NDT is the material/heat-specific transition temperature defined by ASME Code Section III [1], Subsection NB-2300; it is based on a combination of the DW NDT and results from the CVN tests.According to the standard, the temperature NDT = 33.3°C= 13.3°C must be compared with T 68J (=−27.4°C)and T 0.9 mm (=−19.5°C)for the weakest material orientation (TL).Considering these values, it is concluded that RT NDT = NDT = -20°C.

FRACTURE TOUGHNESS TESTS
In Fig. 5, the experimental results of K Jc toughness for LT (Fig. 5a) and TL (Fig. 5b) material orientations are represented as a function of the test temperature T. The reference temperatures T 0 are indicated in the figure (T 0 = -103.04°Cand −97.05°C for the LT and TL orientations, respectively) as well as several confidence bands (0.99, 0.95, 0.5, 0.05, and 0.01).The red points represent censored data (Eq 6) according to ASTM E1921-17a [3].This standard enables the uncertainty in T 0 , ΔT 0 , to be determined; in this case, ΔT 0 is 5.4°C for LT and 6.0°C for TL orientations.
Additionally, the test procedure requires the crack length to be measured for each specimen once the test is completed (provided that an unstable brittle fracture occurred).The measurement for crack length  in precracked Charpy V-notch (PCCv) specimens is shown in Fig. 6a.Moreover, according to ASTM E1921-17a [3], "T 0 characterizes the fracture toughness of ferritic steels that experience onset of cleavage cracking…" and "For any test terminated with no cleavage fracture…the test record is judged to be a nontest, the result of which shall be discarded."For this reason, all fracture surfaces have been examined by means of SEM.Fig. 6b shows a picture in which the border between the fatigue precrack and the region of unstable propagation can be seen.Cleavage was the fracture micromechanism in this region, as demanded by the standard.
In this study, T 0 values were obtained through SEN(B) specimens.As stated in ASTM E1921-17a [3], "there is an expected bias among T 0 values as a function of the standard specimen type […].On average, T 0 values obtained from CT specimens are higher than T 0 values obtained from SE(B) specimens.Best estimate comparison indicates that the average difference between CT and SE(B)-derived T 0 values is approximately 10°C."In this sense, it is important to follow the following recommendation, as given by the standard: "It is therefore strongly recommended that the specimen type be reported along with the derived T 0 value in all reporting, analysis, and discussion of results."

P-T LIMIT CURVES
A P-T curve represents the maximum allowable pressure P for each value of the temperature T considering the presence of a postulated crack in the vessel.Reciprocally, for a specific P, the P-T curve provides the minimum T to guarantee the structural integrity of the vessel.
As stated in Ref. [5], a flaw with depth a equal to one quarter of the thickness t of the vessel (a = t/4, where t = 177 mm for this vessel) must be postulated.P-T curves are required to ensure that the vessel operates within the safety margins regarding brittle failure [27].The previously mentioned combinations of conditions (irradiation level, regulatory scenarios, material description in the DBT region, and crack orientation) were considered in the analysis.In each case, the admissible pressure was obtained as a function of temperature from the failure condition (Eq 10), giving rise to the P-T curves.After obtaining curves corresponding to different combinations of conditions, the most restrictive P-T curve was selected.

P-T Curves in the Unirradiated Condition
Fig. 7a and b shows the P-T curves obtained for (a) the preservice hydrostatic test (SF = 1.5) and (b) transient thermal operations.Each shows three P-T curves that correspond to the material responses in the DBT region dictated by the K IR (T) (Eq 1), K Ic (T) (Eq 2), and MC approach (with P f = 0.05).The area under the curves represents the safety region for operation.In all cases, the most restrictive crack was internal and axially oriented.The inherent overconservatism present in the calculations that are based on regulatory procedures is clearly appreciated.In this sense, note that throughout the complete range of temperatures, the admissible pressure obtained by the MC method is substantially higher than that derived after applying the K Ic or K IR curve.On the other hand, as expected, the most conservative material description comes from the K IR curve.In addition, the hydrostatic test is the least demanding loading situation, while the thermal transient operations of the vessel are the most aggressive.Taking into account that, according to the plant specification, the pressure during a hydrostatic test is 10.86 MPa and the operational pressure of the vessel is 7.24 MPa, the minimum required temperatures to guarantee safety against brittle fracture can be determined.Thus, for the hydrostatic test, the minimum temperatures are 56°C, 20°C, and −43°C, depending on whether the K IR , K Ic , or MC models are employed to describe the material response in the DBT region; the equivalent figures for the in-service hydrostatic test would be 39°C, 6°C, and −64°C.This result invites the reflection that ensuring a temperature of 56°C during a hydrostatic test (as deduced in the case of using the K IR curve) may represent a noteworthy logistic drawback.The corresponding minimum temperatures during the entire transient operations should be 78°C, 40°C, and −21°C, respectively.These results highlight the importance of adopting one or the other approach in the management of the structural assessment of a nuclear vessel.

Influence of Irradiation Embrittlement on P-T Curves
Neutron embrittlement implies the reduction of fracture toughness.The approximately 1:1 correlation between ΔT 41 J and ΔT 0 or ΔRT NDT , as mentioned in the section "The MC" (see Ref. [24]), was considered for obtaining the TTS after irradiation; this information is necessary whether the ASME Code method (curves K Ic (T) and K IR (T) curves) or the MC approach (curve K Jc , Pf = 0.05 (T)) is used.The material condition after 20, 40, and 60 EFPY was determined for structural integrity assessments using the predictions given by "Regulatory Guide 1.99 Rev. 2" [22] and ASTM E900-15e1 [23].
Fig. 8 shows the evolution of the minimum temperature required to avoid brittle fracture as a function of the neutron fluence (expressed in terms of EFPY).The in-service hydrostatic test and thermal transient operations were considered.The most notable conclusions are discussed next: • In each of the figures, the differences in the level of conservatism derived from the selection of the method to describe the DBT zone are clearly observed.K IR (T) is much more conservative than K Ic (T), which, in turn, is substantially more conservative than the MC approach (with P f = 0.05).• It is observed that the effect derived from neutron embrittlement (e.g., a shift in transition temperatures) translates to an increase in the minimum temperatures necessary to avoid brittle failure.This change is noticeable for the first years (from 0 to 20 EFPY), but it gradually attenuates over time; this is a well-known feature of neutron embrittlement.• The results obtained from conventional procedures (K IR and K Ic curves) reveal the drawbacks that could arise when performing the in-service hydrostatic test, since this should be carried out at temperatures of 73°C (K IR ) and 40°C (K Ic ) after 60 EFPY.However, the predictions derived from the MC method imply that there are no difficulties in the execution of this type of tests (the necessary temperatures are below zero, regardless of the level of irradiation).
• Finally, it is observed that the embrittlement derived from the use of the "Regulatory Guide 1.99 Rev. 2" method is greater than that predicted by ASTM E900.Taking into account that ASTM E900 has been developed more recently, its physical foundations are much more robust, and its predictions are more precise, this result can be interpreted as a new source of overconservatism derived from using "Regulatory Guide 1.99 Rev. 2."

MAXIMUM ALLOWABLE CRACK LENGTH
This type of analysis is focused on determining the maximum allowable crack size for the applied pressure (P = 10.86 MPa in the hydrostatic test and P = 7.24 MPa for in-service conditions, which include thermal transient operations) and for the state of embrittlement of the material.As in the case of the P-T curves, different combinations of conditions (irradiation level, regulatory scenarios, material description in the DBT region, and crack orientation) were considered.The assessment of structural integrity is based on the direct comparison between stress intensity factor and material toughness (SF = 1 in Eq 10), which was implemented for the previously mentioned loading conditions (hydrostatic test and thermal transient operations).In addition, two different expressions were considered for the stress intensity factor, namely, the formulations provided by ASME Code Section XI [2] or the FITNET FFS procedure [4].Only axially oriented cracks were considered, since they are more limiting than circumferentially oriented cracks.

Maximum Allowable Crack Length in the Unirradiated Condition
Table 6 summarizes the values of maximum crack size in the internal and external surfaces of the vessel.In all cases, the hydrostatic test, conducted at room temperature (∼25°C), was the limiting scenario.This result contrasts with that obtained in the case of the P-T curves, in which the most restrictive situation corresponded to the thermal transient.This is due to the safety factors adopted in the calculations of the P-T curves, while for the maximum allowable defect, SF = 1 was used.
In general, no substantial differences arise between the results obtained for internal (a int ) or external (a ext ) surface cracks.As in the case of the P-T curves, a marked influence associated with the procedure used to describe the material toughness in the DBT region can be seen.Note that, according to the MC method, admissible cracks range between 45 and 53 % of the thickness of the vessel.In contrast, this range reduces to 22-25 % when using K Ic and 10-12 % when employing K IR .Finally, there is a slight influence associated with the procedure used to define K I ; thus, the formulation proposed by the ASME code leads to results that are slightly more conservative.

Maximum Allowable Crack Length in the Irradiated Condition
For the sake of simplicity, the maximum allowable crack size was determined only after 60 EFPY.Axially oriented cracks were taken into consideration since they are more restrictive.The results are summarized  8 when FITNET FFS was employed.The "Regulatory Guide 1.99 Rev. 2" and ASTM E900 were used to quantify the material embrittlement.In all cases, the loads that are due to hydrostatic tests determine the allowable crack size.This point suggests that a different allowable crack size may be proposed for the hydrostatic test and for operating conditions (thermal transients).Note that the main reduction in maximum crack size is obtained using the "Regulatory Guide 1.99 Rev. 2" [22] predictive model.Furthermore, in most cases, the maximum allowable crack size is shorter than the postulated flaw (a = t/4 = 44.25 mm).In addition, it is observed that the FITNET FFS procedure for K I is slightly less conservative than the ASME Code.

Summary and Conclusions
Neutron embrittlement is the most significant deterioration mechanism of nuclear vessel steels.The reduction undergone by the toughness of the material over the exposure time may influence the viability of the plant and lead to its closure.This phenomenon is of particular relevance when steel operates at a temperature within the DBT region.For this reason, NPPs monitor the evolution of steel properties through so-called surveillance programs, in particular the toughness in the transition region.In spite of this, it is now possible to estimate the embrittlement of the steel using correlations elaborated from experimental and theoretical considerations.
Broadly speaking, two large families of methods are available to assess the structural integrity of a nuclear vessel.First, according to the current legislation, the initial toughness of the steel in the DBT region is described in a deterministic way by means of the initial (unirradiated material) reference temperature RT NDT(U) and its shift over time, ΔRT NDT .It is considered that ΔRT NDT can be obtained experimentally from the results derived from the Charpy impact test so that ΔRT NDT = ΔT 41 J .However, in the absence of experimental data, "Regulatory Guide 1.99 Rev. 2" allows ΔRT NDT to be estimated from the chemical composition of the steel and the neutron fluence.There is a more recent alternative approach based on the description of the behavior of the material in the transition region through the so-called MC, which offers a probabilistic description of the brittle fracture of the material.The reference temperature T 0 is the material property that defines the MC, which is obtained from K Jc fracture toughness tests.In addition, embrittlement can be estimated by means of the procedure of ASTM E900, which takes into account the chemical composition of the material and the neutron fluence while also considering the neutron flux and irradiation temperature.
In this research, an exhaustive study on the performance of the steel (carbon steel grade SA-508M Grade 3 Class 1) of a vessel currently under construction was carried out.The work includes a complete experimental characterization of the material that serves as a basis for the assessment of structural integrity.The main conclusions are summarized as follows: • The characterization comprises the following methods: chemical composition, tension tests (as a function of temperature), Charpy impact tests (as a function of temperature), Pellini DW tests, and K Jc fracture toughness tests (to determine the reference temperature T 0 ).• The high isotropy shown by the material can be highlighted as the most significant result.Thus, the tensile responses in the L and T orientations are virtually identical.This outcome was also observed after comparing the Charpy or fracture toughness tests in the LT and TL orientations, in which TL orientation appears to be slightly weaker (the differences in transition temperatures T 27 J , T 41 J , and T 68 J are between 8.6°C and 12.5°C, and the difference of T 0 amounts to 6°C).In general, the process of forming the steel during the manufacture of the vessel induces microstructural changes that produce the anisotropic behavior of the material [12].In this sense, the strength in the T direction is considered to be lower than in the L direction, and the toughness in the TL orientation is in most cases lower than that in the LT orientation.These experimental results denote, therefore, a careful manufacturing process designed to optimize the strength of the material.• The assessment of structural integrity was carried out from two complementary approaches.On the one hand, the P-T curves were obtained both in the unirradiated condition and for different levels of irradiation: 20, 40, and 60 EFPY.On the other hand, a catalog of maximum admissible defect has been obtained in the same scenarios.• The most notable outcome consists of the inherent overconservatism present in the calculations based on regulatory procedures.The results of the P-T curves or the size of the admissible defect strongly depend on the procedure used to describe the material behavior in the transition region.Thus, when the K Ic or K IR curves are used, a much more conservative picture is obtained than when using the MC method.• Two different scenarios were compared by means of the P-T curves: hydrostatic test and thermal transient conditions (including heating-up and cooling-down operations).The assessment proves that the hydrostatic test is the least demanding loading situation, while the thermal transient operations of the vessel are the most aggressive for plant operation conditions.• The conservatism associated with the methodology of analysis may compromise the viability of the plant.
Thus, the minimum temperature required to perform a hydrostatic test is 56°C, 20°C, and −43°C when the K IR , K Ic , and MC models are employed, respectively.This result is even more severe in the irradiated condition; thus, the hydrostatic test should be carried out at temperatures above 80°C (K IR ) or 40°C (K Ic ) after 60 EFPY.In contrast, the minimum temperature that arises after employing the MC method is below zero in all cases.• The maximum size of admissible defect is also affected by the procedure that is used to characterize the DBT region.In this sense, the admissible cracks range between 45 % and 53 % of the thickness of the vessel when using the MC.In contrast, this range reduces to 22-25 % when using K Ic and 10-12 % when employing K IR .
Regulatory authorities in some countries have already incorporated the MC procedure as an available tool for surveillance programs and for the assessment of the structural integrity of nuclear vessels.The conclusions reached in this study show that the methods traditionally used for the description of fracture toughness in the DBT region suffer from intrinsic limitations that encourage the modification of the regulations to fully incorporate the MC approach.

FIG. 1
FIG. 1 Distribution of specimens in the sector of the surveillance ring used in this study.
Charpy toughness in the DBT region in the LT and TL orientations.LT 12 TL 15 Characterization of the K Jc toughness in the DBT region in the LT and TL orientations to determine T 0 tensile properties in the L and T orientations for a wide range of temperatures (from −145°C to 300°C).

FIG. 3
FIG. 3 Influence of temperature on yield stress and tensile strength: (a) L orientation and (b) T orientation.

FIG. 4
FIG. 4 Results of the CVN tests in the LT and TL orientations: (a) absorbed energy versus temperature, (b) LE versus temperature, and (c) SFA versus temperature.

FIG. 7
FIG. 7 P-T curves obtained for (a) a preservice hydrostatic test and (b) transient thermal operations, core critical.

TABLE 1
Description of the experimental scope.

TABLE 2
[25]ical composition (wt.%) of the steel (mean and standard deviation) and comparison with the limits of the ASME Code[25].
FERREN ˜O ET AL.ON NUCLEAR VESSEL INTEGRITY Journal of Testing and Evaluation Copyright by ASTM Int'l (all rights reserved); Fri Sep 18 04:49:27 EDT 2020 Downloaded/printed by Universidad De Cantabria Biblioteca (Universidad De Cantabria Biblioteca) pursuant to License Agreement.No further reproductions authorized.

TABLE 3
Parameters of the exponential fitting of the stress-temperature curves.

TABLE 4
Relevant results derived from the CVN impact test.

TABLE 5
Experimental results obtained from the Pellini DW test.

TABLE 6
Results of the maximum allowable crack length (internal and external surfaces of the vessel) in the unirradiated condition.

TABLE 7
Results of the maximum allowable crack length (internal and external surfaces of the vessel) after 60 EFPY.K I is obtained by means of the ASME Code.˜O ET AL.ON NUCLEAR VESSEL INTEGRITY in Table 7 using the expression for K I proposed by the ASME Code and in Table FERREN

TABLE 8
Results of the maximum allowable crack length (internal and external surfaces of the vessel) after 60 EFPY.K I is obtained by means of the FITNET FFS.