A

ytRsBTC©

1

nlpf

AT(detASNRmMRcPU

0d

Nuclear Engineering and Design 237 (2007) 2098–2104

Reactor pressure vessel embrittlement of NPP borssele:Design lifetime and lifetime extension

F.J. Blom ∗NRG, P.O. Box 25, NL-1755 ZG Petten, The Netherlands

Received 9 May 2006; received in revised form 10 February 2007; accepted 13 February 2007

bstract

Embrittlement of the reactor pressure vessel of the Borssele nuclear power plant has been investigated taking account of the design lifetime of 40ears and considering 20 years subsequent lifetime extension. The paper presents the current licensing status based on considerations of materialest data and of US nuclear regulatory standards. Embrittlement status is also evaluated against German and French nuclear safety standards.

esults from previous fracture toughness and Charpy tests are investigated by means of the Master curve toughness transition approach. Finally,

tate of the art insights are investigated by means of literature research. Regarding the embrittlement status of the reactor pressure vessel oforssele nuclear power plant it is concluded that there is a profound basis for the current license up to the original end of the design life in 2013.he embrittlement temperature changes only slightly with respect to the acceptance criterion adopted postulating further operation up to 2033.ontinued safe operation and further lifetime extension are therefore not restricted by reactor pressure vessel embrittlement.

1

2007 Elsevier B.V. All rights reserved.

. Introduction

The status of the reactor pressure vessel of the Borssele

uclear power plant shows no concern for operation to the designifetime of 2013. The interest of this paper for Borssele is theossibility of lifetime extension beyond 2013. Therefore, theollowing goal is set for this study.

Abbreviations: AMES, Ageing of Materials and Evaluation Studies; ASME,merican Society of Mechanical Engineers; ASTM, American Society foresting and Materials; CF, Chemistry Factor given in tables in RG1.99 rev. 21988); CFR, Code of Federal Regulations; Cu, copper content; EDF, Electricitee France; φ, Fluence given in 1019 n/cm2; FIM, French Average Irradiationmbrittlement prediction formula; FIS, French Upper Bound Irradiation embrit-lement prediction formula; HAZ, Heat Affected Zone; IAEA, Internationaltomic Energy Agency; KTA, Kerntechnischer Ausschuss (German Nuclearafety Standards Commission); Ni, nickel content; NPP, Nuclear Power Plant;RC, Nuclear Regulatory Commission; NUREG, Report series from Nuclearegulatory Commission (NRC); P, phosphorus content; PTS, pressurized ther-al shock; RCC-M, Regles de Conception et de Construction relatif auxateriels mecaniques (French nuclear material code); RG, Regulatory Guide;SEM, Rules for in-service surveillance of mechanical equipment (Frenchode); RTNDT, Nil Ductility Reference Temperature; SOP, “Staal Onderzoeksrogramma” (Material surveillance program); USE, Upper Shelf Energy; US,nited States; WOL, wedge opening loading∗ Tel.: +31 224 56 8186; fax: +31 224 56 8490.

E-mail address: blom@nrg-nl.com.

dnmo

2

2

tt(Tmocssoz

029-5493/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.nucengdes.2007.02.008

.1. Goal

Review the state-of-the-art insights about reactor vessel irra-iation embrittlement and safety standards concerning Borsseleuclear power plant. The review will result in an extensiveulti-disciplinary irradiation embrittlement survey for long term

peration of the Borssele reactor pressure vessel.

. Current status

.1. RTNDT

The current status of the reactor pressure vessel with respecto irradiation embrittlement is based on the measured transitionemperatures RTNDT out of three material surveillance programs“Staal Onderzoeks Programma’s”, SOP-0, SOP-1 and SOP-2).he transition temperatures are determined by Charpy energyeasurements on the unirradiated material (SOP-0) and two sets

f irradiated samples (SOP-1 and SOP-2) placed closer to theore than the actual vessel. These surveillance programs repre-

ent the irradiation embrittlement status of the vessel at a latertage by means of an elevated dose. The most irradiated partsf the vessel are ring 3, ring 4, weld W03 and the heat affectedone (HAZ) of this weld (see Fig. 1). The measured RTNDT for

F.J. Blom / Nuclear Engineering and

Fig. 1. Reactor pressure vessel of the Borssele nuclear power plant.

Table 1RTNDT (◦C) SOP-0 to SOP-2 at different locations

Location RTNDT (SOP-0) RTNDT (SOP-1) RTNDT (SOP-2)

Ring 3 −10 −0.3 4.2Ring 4 −20 −3.3 3.1WH

th

aro(ci91ti

TD

SS

2

nU1cCltoR(

R

wgpr

�

wR(φ

i

st

twbostcRo

2

NA

eld W03 −45 −13.1 −4.1AZ −5 7.5 27.4

he different zones are shown in Table 1. It can be seen that theighest RTNDT is found in the HAZ and is equal to 27.4 ◦C.

Subsequently, the lead factors are determined for SOP-1nd SOP-2. These are used to determine the year, which isepresented by both programs. These calculations are basedn calculations using Point kernel methods and Twotran S2Oosterkamp and Dufour, 1983). The represented year is cal-ulated using the actual load factor. The cumulative load factors extrapolated to 2033 using a very high actual load factor of

7%. The transformation of the core to a low-leakage one after0 years of operation is taken into account in order to determinehe representative year for SOP-1 and SOP-2, which are shownn Table 2.

able 2ating result for SOP-1 and SOP-2

Year

OP-1 2000OP-2 2012

tBtrTvlfli(o

Design 237 (2007) 2098–2104 2099

.2. Prediction equation RG1.99 rev. 2

From a conservative viewpoint, the licensing of Borsseleuclear power plant is based on the maximum RTNDT from theS Regulatory Guide 1.99 rev. 1 (RG1.99 rev. 1, 1977) and0CFR section 50.61 (10CFR 50.61, 1984). These standardsonservatively predict the RTNDT based on the initial RTNDT,u and Ni content and fluence from correlation relations in

arge series of reactor pressure vessel test data. These predic-ion equations give a RTNDT of 44 ◦C in the HAZ after 40 yearsf operation. The updated version of the prediction is given inG1.99 rev. 2 (RG1.99 rev. 2, 1988) and 10CFR section 50.61

10CFR 50.61, 2003).The equation is given below,

TNDT = Initial RTNDT + �RTNDT + margin (1)

here InitialRTNDT is the initial RTNDT (see Table 1). Mar-in is equal to twice the standard deviation for the surveillancerograms investigated (31 ◦C for welds and 19 ◦C for other mate-ials). �RTNDT is given below,

RTNDT = (CF)φ(0.28−0.10 log φ) (2)

here CF is a ‘Chemistry Factor’, which is given in tables inG1.99 rev. 2 (RG1.99 rev. 2, 1988) and 10CFR section 50.61

10CFR 50.61, 2003) as a function of copper and nickel content.is the fluence given in 1019 n/cm2 (>1 MeV). RTNDT is given

n ◦F due to the US background of the documents.The RTNDT for the different locations, calculated by these

tandards, is given in Fig. 2. The maximum RTNDT is found inhe HAZ and is equal to 45 ◦C in 2012 and 48 ◦C in 2033.

According to 10CFR section 50.61 (10CFR 50.61, 2003)here is a PTS screening criterion for the maximum RTNDT,hich has to remain below 149 ◦C in welds. The criterion isased on probabilistic PTS analyses performed on a databasef US reactor pressure vessels (see RG1.154, 1987). If the PTScreening criterion is met, the probability of through wall frac-ure is lower than 5 × 10−6 per reactor year. Fig. 2 shows that theriterion is amply met. Therefore, it can be concluded that theTNDT criteria from (10CFR 50.61, 2003) do not pose a threatn further operation of the reactor pressure vessel until 2033.

.3. Upper shelf energy

The upper shelf energy (USE) demand is determined by USRC regulations in (10CFR 50.60, 1996) and (10CFR part 50ppendix G, 2003b). These regulations state that the USE has

o remain above 68J. The initial USE for the pressure vessel oforssele is higher than 200 J for all locations. The decrease of

he USE can be determined by means of a graph from (RG1.99ev. 2, 1988) depending on the Cu and Ni content and fluence.his prediction is also based on a large series of reactor pressureessel test data. Since the Cu content of the pressure vessel isow, the decrease of the USE is low and limited to 31% at a

uence of 3.5 × 1019 n/cm2. The measured decrease of the USE

s limited to 27%. It can be concluded that the USE criteria from10CFR part 50 Appendix G, 2003b) do not limit the operationf the pressure vessel.

2100 F.J. Blom / Nuclear Engineering and Design 237 (2007) 2098–2104

for (R

3

ptca1iv

3

osmtgea

m(Asa

p(G

3

id

Fpsiv

�

wgtiaFttf(

sA

�

Tilth

d

Fig. 2. RTNDT as function of year

. International safety standards

In this chapter, the adopted US safety standards are com-ared to the French RCC-M and German KTA standards in ordero place the currently used standards in perspective. Generalomparisons between different international safety standardsre given in IAEA (1999) and AMES reports no. 4 (Gerard,995) and 6 (Petrequin, 1996). In this chapter, the compar-sons will be made specifically for the Borssele reactor pressureessel.

.1. Experimental determination of RTNDT

The transition temperature RTNDT is determined by meansf ASME III, NB-2331 (2001) according to the US regulatorytandard (10CFR part 50 Appendix G, 2003b). In brief, thiseans that the RTNDT is determined by means of the Charpy

est and satisfies the following conditions: at a temperature notreater than RTNDT+33 ◦C, each specimen of the Cv test shallxhibit at least 0.89 mm lateral expansion and not less than 68 Jbsorbed energy.

According to the French regulatory standards RTNDT is deter-ined by means of the RCC-M code Section 2, article MC1240

RCC-M, MC-1240, 1993a,b)). The conditions correspond toSME III, NB-2331 (ASME III, NB-2331, 2001). The German

tandard for determination of RTNDT, KTA 3201.2 (KTA, 1996),lso corresponds to (ASME III, NB-2331, 2001).

The transition temperature RTNDT for the Borssele reactorressure vessel is determined according to ASME III, NB-23312001). Concluding from the above this satisfies US, French anderman standards.

.2. Prediction equations (RTNDT)

The prediction equation for the adopted US standards is givenn RG1.99 rev. 2 (1988) and discussed in Section 2.2. The pre-icted RTNDT graph is given in Fig. 2.

3om3

G1.99 rev. 2, 1988) and SOP-1/2.

The French regulatory standards use the prediction equationsIS and FIM from RSEM article B-7212 (RSEM, 1990). Therediction is based on research by Brillaud et al. (1987), where aeries of French surveillance data is analysed. The FIM equations based on the mean data and FIS is based on the upper boundalues. �RTNDT according to FIS is given by,

RTNDT(FIS) = 8+(24 + 1537(P − 0.008)+238(Cu − 0.08)

+191Ni2Cu)φ0.35 (3)

here φ is the fluence in 1019 n/cm2 (>1 MeV) and RTNDT isiven in ◦C. The difference between these French equations andhe one given in RG1.99 rev. 2 is the addition of phosphorus (P)n FIS and FIM and the combined influence factor of nickel (Ni)nd copper (Cu). Also the fluence exponent is slightly higher inIS and FIM. The graph for FIS is shown in Fig. 3. Here, only

he FIS results are shown, since the FIM results show the samerend and are slightly lower and less conservative. The valuesor FIS are comparable to the ones obtained from RG1.99 rev. 21988).

Besides the FIS and FIM equations, the French regulatorytandards use the equation from RCC-M Appendix ZG (RCC-Mppendix ZG, 1993b),

RTNDT(RCC-M) = (22 + 556(Cu − 0.08)

+2778(P − 0.008))φ0.5 (4)

he results are shown in Fig. 4. The figure shows the largencrease of the RTNDT of the weld W03. This is caused by thearger copper content of the weld, which has a larger contribu-ion in the RCC-M equation. However, the HAZ remains theighest absolute value.

The German regulatory standards have abandoned the pre-iction equations, as from the last version in 2001 of KTA

203 (KTA, 2001). This German standard prescribes the usef surveillance results only. The RTNDT is bound to a maxi-um value depending on fluence, which is presented in KTA

203 (KTA, 2001) as RTlimit. This limit is more severe than

F.J. Blom / Nuclear Engineering and Design 237 (2007) 2098–2104 2101

of y

t(bisT

3

itsdmn

mpr

3

3

dua

Fig. 3. RTNDT as function

he US limit from 10CFR section 50.61 (10CFR 50.61, 2003)RTNDT<149 ◦C). The results from the surveillance program cane interpolated or extrapolated according to �RTNDT = Aφn. Thenterpolated function using a conservative value of n = 0.5 ishow in Fig. 5 together with the KTA limit function for RTNDT.his figure shows that RTNDT remains below the limit curve.

.3. Upper shelf energy

The conditions on the Upper Shelf Energy (USE) are accord-ng to 10CFR section 50.60 (10CFR section 50.60, 1996) boundo a minimum of 68 J. The same limit is given by the German

tandard KTA 3203 (KTA, 2001). The French regulatory stan-ards do not prescribe a minimum value of USE. In practice, ainimum value of 56 J is used according to the AMES report

o 4. (Gerard, 1995).

crTs

Fig. 4. RTNDT as a function of yea

ear for FIS and SOP-1/2.

For the Borssele pressure vessel the minimum USE of 68 J isaintained, as described in Section 2.3. Therefore, the reactor

ressure vessel is in accordance with the French and Germanegulatory standards with respect to USE.

.4. Surveillance programs

.4.1. Withdrawal scheduleThe US 10CFR section 50.60 (10CFR section 50.60, 1996)

oes not prescribe general demands on the withdrawal sched-le of the samples, it has to be submitted and approved by theuthorities (NRC). The French regulatory standard RSEM arti-

le B-7213 (RSEM, 1990) prescribes more than one surveillanceesult, where the last one has to be equivalent to the end of life.he German regulatory standard KTA 3203 (KTA, 2001) pre-cribes a minimum of two surveillance results. The first one has

r for RCC-M and SOP-1/2.

2102 F.J. Blom / Nuclear Engineering and Design 237 (2007) 2098–2104

Fig. 5. RTNDT as a function of year for KT

Table 3T0 for SOP-1/2 at different locations [◦C]

Ring 3 Ring 4 Weld W03

SS

tavs

3

tfetrpn1lI(t2

lut

4

TaOnt

MoSv2s−nTa

TC

L

RRWH

OP-1 −78.7 −81.5 −10.4OP-2 −66.8 −62.8 −38.2

o be situated around 50% of the lifetime and the second onet least at 100% of the lifetime. Therefore, the reactor pressureessel is in accordance with the French and German regulatorytandards until 2012 (see Table 2).

.4.2. Lead factorThe lead factor is the factor between the fluence received by

he samples and the fluence received by the vessel wall. Theseactors are determined in Oosterkamp and Dufour (1983) andqual to 9.5 for SOP-1 and 6 for SOP-2. A resume of the interna-ional standards with respect to the lead factor is given in AMESeport no. 4 (Gerard, 1995). The US regulatory standard 10CFRart 50 Appendix H (10CFR part 50 Appendix H, 2003c) doesot limit the value of the lead factor. According to IAEA tecdoc120 (IAEA, 1999) and AMES report no. 4 (Gerard, 1995), theead factor for the French surveillance programs is limited to 3.

t is recognised that these documents are not particularly recent1999 and 1995) and it is understood that EDF no longer applieshis limit. The German regulatory standard KTA 3203 (KTA,001) requires that the lead factor lies between 1.5 and 12.

et

t

able 4omparison T0 to RTNDT [◦C] for SOP-1/2

ocation RTT0 (SOP-1) RTNDT (SOP-1) � (T0−NDT) (SOP

ing 3 −59.3 −0.3 −59.0ing 4 −62.1 −3.3 −58.8eld W03 9.0 −13.1 22.1AZ x 7.5 x

A limit and SOP-1/2 (interpolated).

It can be concluded that the Borssele reactor pressure vesselead factor satisfies US and German regulatory standards. It isnderstood that the Borssele lead factor does not conflict withhe present French requirements.

. Master Curve

In this section, the Master Curve analyses are discussed.hese analyses are performed on the test results from SOP-1nd SOP-2. The tests are performed on WOL samples (Wedgepening Loading). These tests determine the fracture tough-ess directly, instead of the absorbed energy measurements ofhe Charpy test.

Subsequently, the test results are analysed by means of theaster Curve method. A Master Curve is constructed by means

f the ASTM 1921-02 (ASTM 1921-2, 2002b). The data fromOP-1 and SOP-2 do not contain sufficient data to construct aalid Master Curve according to ASTM 1921-02 (ASTM 1921-, 2002a,b). This standard states that there have to be 8 validamples is the lower end of the curve (T − T0) between −50 and36 ◦C. For the test results from SOP-1 and SOP-2, a maximum

umber of 3 valid results is obtained. The reference temperature0 obtained is given in Table 3. For the HAZ no WOL specimensre available. For SOP-0 only KIc values have been measured at

xtreme low temperatures (−196 to −150 ◦C), which is too lowo obtain any result for a Master Curve.

Since the results from the Master Curve analysis are not valid,hey cannot be used for a safety demonstration. However, the

-1) RTT0 (SOP-2) RTNDT (SOP-2) �(T0−NDT) (SOP-2)

−47.4 4.1 −51.5−43.4 3.1 −46.5−18.8 −4.1 −14.7

x 27.4 x

F.J. Blom / Nuclear Engineering and Design 237 (2007) 2098–2104 2103

Table 5Comparison of RTNDT (HAZ) using different prediction equations

Year SOP-1 SOP-2 RG1.99 FIS RCC-M NUREG/CR-6551 ASTM E900-02

2000 7.5 42.8 41.2 33.1 36.6 35.12012 27.4 45.2 46.8 41.2 40.1 38.52013 45.4 47.4 42.2 40.6 39.02 .4

rfNiiTtgwsr

5

5

gm1ki0ai

eawscp

6

soslaadatbs

prettee

R

A

A

A

A

B

E

E

G

I

K

K

O

P

P

P

P

P

033 47.7 54

esults can be compared to the Charpy results in order to checkor discrepancies. Therefore, ASME code case N-629 (ASME-629, 1999) is used to translate the T0 in a value RTT0, which

s comparable to RTNDT. The relation between RTT0 and T0s given by: RTT0=T0+19.4 ◦C. The comparison is shown inable 4. It can be seen that the Master Curve results are lower

han the Charpy results. The only case where the Master Curveives a higher reference temperature is the weld W03 for SOP-1,hich is attributed to a non-homogeneous material. The results

how that there are no discrepancies between the Master Curveesults and the Charpy ones.

. State of the art

.1. Prediction equations (RTNDT)

The database with test data used to construct the equationiven in RG1.99 rev. 2 (1988) is further extended to give aore updated correlation in NUREG/CR-6551 (Eason et al.,

998). This correlation incorporates more mechanistic materialnowledge. The prediction equation is then further evolved andncorporated into the ASTM standard E900-02 (ASTM E900-2, 2002a). The relevance of this equation is increasing, sinceccording to English and Hyde (2003) this equation can be usedn the successor of RG1.99 rev. 2 (1988).

The RTNDT values calculated by the different predictionquations for the HAZ of the Borssele reactor pressure vesselre compared in Table 5. All equations show a very low RTNDTith respect to the limit given by Proposed rule making 10CFR

ection 50.61 (2003) (RTNDT < 149 ◦C). Therefore, it can be con-luded that the irradiation embrittlement of the Borssele reactorressure vessel is minor.

. Conclusions

Embrittlement of the reactor pressure vessel of the Bors-ele nuclear power plant has been investigated taking accountf the design lifetime of 40 years and considering 20 yearsubsequent lifetime extension. The paper presents the currenticensing status based on considerations of material test datand of US nuclear regulatory standards. Embrittlement status islso evaluated against German and French nuclear safety stan-ards. Results from previous fracture toughness and Charpy tests

re investigated by means of the Master curve toughness transi-ion approach. Finally, state of the art insights are investigatedy means of literature research. Regarding the embrittlementtatus of the reactor pressure vessel of Borssele nuclear power

P

R

53.1 44.7 42.8

lant it is concluded that there is a profound basis for the cur-ent license upto the original end of the design life in 2013. Thembrittlement temperature changes only slightly with respecto the acceptance criterion adopted postulating further opera-ion up to 2033. Continued safe operation and further lifetimextension are therefore not restricted by reactor pressure vesselmbrittlement.

eferences

SME Code Case N-629, Use of Fracture Toughness Test Data to EstablishReference Temperature for Pressure Retaining Materials Section XI Division1, 1999.

SME, 2001 section III, NB-2331, Test requirements and acceptance standards,Materials for Vessels, 2001.

STM, 2002a. E900-02, Standard guide for predicting radiation induced tran-sition temperature shift in reactor vessel materials, E706 (IIF).

STM, 2002b. E1921-02, Standard test method for determination of referencetemperature, T0, for ferritic steels in the transition range.

rillaud, C., Hedin, F., Houssin, B., 1987. A comparison between French surveil-lance program results and predictions of irradiation embrittlement, Influenceof radiation on material properties. In: 13th International Symposium ASTMSTP 956, pp. 420–447.

ason, E.D., Wright, J.E., Odette, G.R., 1998. Improved Embrittlement Corre-lations for Reactor Pressure Vessel Steels, NUREG/CR-6551.

nglish, C.A., Hyde, J.M., 2003. Radiation embrittlement of reactor pressurevessel steels. In: Comprehensive Structural Integrity. Elsevier Science (chap-ter 6.08).

erard, R., 1995. Survey of National Regulatory Requirements, AMES ReportNo. 4, EUR 16305 EN.

AEA-TECDOC-1120, 1999. Assessment and management of ageing of majornuclear power plant components important to safety: PWR pressure vessels.

TA 3201.2, 1996. Components of the Reactor Coolant Pressure Boundary onLight Water Reactors, Part 2: Design and Analysis.

TA 3203, 2001. Surveillance of the Irradiation Behaviour of Reactor PressureVessel Materials of LWR Facilities.

osterkamp, W.J., Dufour, L.B., 1983. Neutron embrittlement of the reactorvessel in Borssele as determined from Charpy specimens, Kema scientific& technical reports, vol. 1, no. 4, pp. 45–54.

etrequin A., 1996. A review of formulas for predicting irradiation embrittlementof reactor vessel materials, AMES Report no. 6, EUR 16455 EN.

roposed Rule Making, 1996. 10 CFR Part 50. Section 50.60, Acceptance criteriafor fracture prevention measures for lightwater nuclear power reactors fornormal operation, NRC.

roposed Rule Making, 1984. 10 CFR Part 50. Section 50.61, Fracture Tough-ness Requirements for Protection Against Pressurized Thermal Shock,NRC.

roposed Rule Making, 2003. 10 CFR Part 50. Section 50.61, Fracture Tough-ness Requirements for Protection Against Pressurized Thermal Shock, NRC.

roposed Rule Making, 2003b. 10 CFR Part 50. Appendix G. Fracture toughness

requirements, NRC.

roposed Rule Making 10 CFR part 50, 2003c. Appendix H, Reactor vesselmaterial surveillance program requirements, NRC.

CC-M section III article MC1240, 1993a. Determination of reference nil duc-tility transition temperature.

2 and

RR

R

104 F.J. Blom / Nuclear Engineering

CC-M Subsection Z, Appendix ZG 3430, 1993b. Irradiation effects.eg. Guide 1.99, 1997. Radiation embrittlement of reactor vessel materials,

rev. 1.eg. Guide 1.99, 1988. Radiation embrittlement of reactor vessel materials,

rev. 2.

R

R

Design 237 (2007) 2098–2104

egulatory Guide 1.154, 1987. Format and content of plant-specific pres-surized thermal shock safety analysis reports for pressurized waterreactors.

SEM, 1990. In service inspection rules for mechanical equipment of PWRnuclear islands (in French), article B-7212.