Annealing for plant life management

Int. J. Pres. Ves. & Piping 60 (1994) 217-222 ~) 1994Elsevier ScienceLimited Printed in Northern Ireland. All fights reserved 0308-0161194l$07.00

ELSEVIER

Annealing for plant life management: Hardness, tensile and Charpy toughness properties of irradiated, annealed and re-irradiated mock-up low alloy nuclear pressure vessel steel Philip Tipping & Robin Cripps Paul Scherrer Institute, Wiirenlingen and Villigen, CH-5232 Villigen PSI, Switzerland

(Received 20 October 1993; accepted 2 November 1993)

Hardness, tensile and Charpy properties of an irradiated (I) and irradiatedannealed-reirradiated (IAR) mock-up pressure vessel steel are presented. Spectrum tailored pressurized light water reactor (PWR) irradiation at 290°C by fast neutrons up to nominal fluences of 5 x 1019/cm2 (E ~ 1 MeV) in a swimming pool type reactor caused the hardness, tensile yield stress and tensile strength to increase. Embrittlement also occurred as indicated by Charpy toughness tests. The optimum annealing heat treatment for the main program was determined using isochronal and isothermal runs on the material and measuring the Vickers microhardness. The response to an intermediate annealing treatment (460~C for 18 h), when 50% of the target fluence had been reached and then irradiating to the required end fluence (IAR condition) was then monitored further by Charpy and tensile mechanical properties. Annealing was beneficial in mitigating overall hardening or embrittlement effects. The rate of reembrittlement after annealing and re-irradiating was no faster than when no annealing had been perfomed. Annealing temperatures below 440°C were indicated as requiring relatively long times, i.e. ->168h to achieve some reduction in radiation induced hardness for example. The overall benefit of an appropriate annealing treatment has been demonstrated on a laboratory scale, and annealing is therefore indicated as being a useful tool in the overall strategy for achieving plant life management goals.

by adjusting the fuel bundle configuration or by using dummy stainless steel elements in the core for example. This type of core management simply delays the point in time when the PV EOL fast neutron fluence, e.g. 1-2 × 10~9/cm2 will be reached, thus potentially extending the NPP Life. Annealing the PV may also be a way • to achieve plant life management (PLM) goals due to regenerating mechanical properties which have been affected by neutron irradiation. The following work describes how the annealing parameters of time and temperature may be selected to obtain recovery in mechanical

1 INTRODUCTION As nuclear power plants (NPP) become older and approach their nominal end of life (EOL) and orders for new plants do not keep up with the potential loss of electricity production, it is essential to manage the existing plants in order to ensure delivery of energy at the highest levels of safety. If deterioration in the mechanical properties of the pressure vessel (PV) (due to neutron irradiation) is solely governing the overall life of the NPP, then it is possible to reduce the rate of neutron fluence accumulation 217

218

Philip Tipping, Robin Cripps

properties. Information and data are provided which indicate that for these experiments, material and conditions, the rate of re-hardening (or embrittlement) after annealing and reirradiating was no faster than when no annealing had been performed.

2 EXPERIMENTAL PROCEDURES 2.1 Material The rolled plate of mock-up low alloy ferritic PV steel (similar to A533-B) was normalized at 900°C, quenched at 880°C, tempered at 665°C for 12 h and finally stress releived at 620°C for 40 h. Due to the intentionally selected high copper content of 0.14wt% and a nickel content of 0.84wt%, this 0.18wt% carbon mock-up PV steel can be placed into the category of being radiation sensitive.

2.2 Irradiation The specimens (Charpy, tensile and hardness coupons) were loaded into instrumented stainless steel capsules and placed in a swimming pool type materials testing reactor (MTR) where they received a neutron spectrum similar to that in a pressurized water reactor (PWR) at a flux rate of about 5 x 1012/cm2/s. Nominal fluences up to 5 × 1019/cm 2, were accumulated at 290°C with a lead factor of about 100. Irradiation times lasted for up to 4 months according to the desired fluence. It should be noted that all fluences quoted hereafter are those neutrons having an energy ->1 MeV. Dosimetric fluences were obtained from iron, nickel, niobium and copper monitors which were incorporated in the capsules. The temperature (+5°C) was achieved by gamma heating and controlled using heliumnitrogen mixtures in the gap between the capsule wall and the specimens. Some experiments were done by allowing the specimens to accumulate approximately 50% of their foreseen target fluences and then applying a heat treatment (460°C x 18 h) to them whilst still in the capsule. The heat treatment had been determined previously using isochronal and isothermal hardness response analysis (see Section 2.3 below). After heat treatment, the specimens were re-irradiated to the required fluence targets, giving the I A R condition. This enabled an

assessment of the rate of re-hardening (embrittlement) after annealing which is an important aspect for the practical case.

2.3 Determination of annealing parameters Coupons, 4 x 4 × 2 m m , having a fluence of 2-23x 1019/cm 2 were heated under helium in 30min isochronal runs; the room temperature microhardness (HV0.2) was measured. Only at temperatures ->400°C did the Vicker's microhardness (HV0.2) begin to recover; by 500°C the hardness had reached the unirradiated (U) value (Fig. 1). Correspondingly, a series of isothermal runs were performed between 440°C and 500°C and a treatment of 460°C x 18h was selected because it caused approximately 90% relative recovery in the microhardness in a fairly practical time, Fig. 2.

2.4 Tensile testing A servohydraulic machine installed in a Hotcell was used in the extension controlled mode; the strain rate on the specimens was 6 x 10-6/s. For temperatures higher than ambient, electric heaters were used assuring an accuracy of +5°C. The tensile specimens (taken from standard Charpy size blanks at the 1/4 or 3/4 thickness of the original plate) had their axes parallel to the rolling direction.

300

I - FLUENCE

280

O > "I" 09 09 LU Z a n" ,,~ "tO n,-'

=

ISOCHRONAL

260

F

2 - 2 3 x 1 0 1 9 / c m 2, E > I M e V . TIME =

30 min

I

240

_o

220

200 0

'

'

'100"

" '200'

' '300'

TEMPERATURE

' '400'

' '500'

' '600

[°C]

Fig. 1. Isochronal behaviour of the irradiated mock-up A533-B type steel. Softening of the material commenced only after 400°C was exceeded; by 500°C the original hardness was reached.

Annealing for plant life management 100

:

~ ,.~

t± _~

•-'_: -':

8O

i 7060500(/7480C//440C /i/, /

504030

I I

10 0

//x//9,

~ / ~,~_ /~,, I'

~'

'

18~h

219

recovery in this material having a fluence of 2.23 × 1019/cm2. The time of 18 h was selected as a minimum to achieve adequate recovery of hardness; an extended time, up to 168 h (1 week) would likely also produce similar recovery levels; extrapolation of the isotherms to 168 h (Fig. 2) certainly indicates this behaviour. The present annealing parameters therefore lie within the usually quoted and found temperature range for A533-B type steel irradiated at 290°C. 1

3.2 Tensile and Charpy properties I-FLUENCE= 2.23 x l_l_cm 2 , n 9 / E > 1MeV

' '"'1

. . . . . . . .

lo

!

loo

. . . . . . . .

I

looo

. . . . . . . .

1000o

TIME [MIN] Fig. 2. Isothermal behaviour. The choice of an optimum

annealing temperaturedependson practicalities;500°C causes rapid recoverywhilst temperatures below440°C are likely to lead to long times. At least 90% recovery in hardness is to be expectedusing460°Cx 18h.

2.5 Charpy testing The standard Charpy testing procedure was used. The hammer (ISO-V tup) struck the specimens at 5m/s. Specimen temperatures were obtained using mixtures of dry ice and alcohol or an integral instrumented oven and pneumatically operated fast transfer device for temperatures greater than 20°C. The error in temperature was reduced by overheating or undercooling according to previously determined calibrations. The Charpy specimens, taken from the 1/4 and 3/4 layers in the original plate, had the T-L orientation.

3 RESULTS AND DISCUSSION 3.1 Heat treatment The 30 min isochronal behaviour indicated where the annealing, as defined by decreasing hardness, began to take place. This temperature was around 400°C, indicating that at least ll0°C more than the irradiation temperature was required under these isochronal conditions (Fig. 1). The amount of hardness recovery increased as the temperature increased, until at about 500°C full hardness recovery (to the unirradiated value) was obtained. A series of isothermal tests confirmed that the general temperature range between 440 and 500°C was practical for causing hardness

T h e absolute changes in tensile yield stress and strength data for the I and IAR conditions at 20°C, 150°C and 300°C as a function of neutron fluence are presented in Figs 3-5 respectively. The benefit of annealing at 460°C for 18 h at 50% of the target fluence is evident in that the IAR data exhibit less change in property compared to the data obtained for the I condition for the same fluence level. The rate of re-hardening for IAR appears to be no more than that observed for the I material state for all temperatures investigated. This information is of importance for utilities considering the annealing option for PLM because an accelerated embrittlement after annealing is counter-productive. The tensile yield stress and strength behaviour follow the same general pattern of hardening and embrittlement and mitigation for each of the test temperatures used. An explanation for the similar rate of change in the tensile properties after annealing and re-irradiating compared to the irradiated only trends is that this chosen annealing treatment had caused mainly a reduction in .irradiation induced point defects and perhaps only some coarsening of copper-rich particles. Since the IAR behaviour as a function of fluence could be described globally by a vertical downwards shift (less property change) compared to the I condition at the same fluences, the re-embrittling mechansim for the IAR condition could be phenomenologically the same as for the I condition. The advantage the IAR condition has over the I one is that it has probably less (or modified) irradiation damage and possibly some overaged copper-rich particles compared to the I condition at the same fluence level; more work, out of the scope of the present paper (e.g. TEM and SANS investigations) is indicated here. Beneficial effects of annealing were also found for the Charpy notch ductility (toughness)

Philip Tipping, Robin Cripps

220

YIELD STRESS I

!:t

t

TENSILESTRENGTH

I

I

flit

..

! ° -20°l

I 1

I 2

I

I

120

" "/"I

"~''

/

;

t

/

/

/

/

//

i

//

I I I 3 4 1 FLUENCE[/101~cm2 ]

I

I

F

2

3

4

5

Fig. 3. Absolute changes in tensile properties at 20°C. The hardening/embrittlement trends at equivalent fiuences for the irradiated and irradiated-annealed-re-irradiated conditions indicate the benefit of annealing.

properties (Table 1). Not only were the transition temperature shifts lowered but also the upper shelf energy (USE) was restored somewhat for the IAR condition. This is a positive indication for utilities which may be approaching pressure/ temperature constraints due to a lowered upper shelf energy. The unirradiated 41 J and 68J energy levels were present at -23°C and - l l ° C respectively, whilst the upper shelf energy was 200 J. In this experiment, the 41 J and 68 J ductile to brittle transition temperature shifts were reduced from typically 90°C to 60°C by the annealing schedule applied. Likewise, the upper shelf energies were restored from 146 J to about

175 J, i.e. approaching the unirradiated value of 200J. (N.B. Data refer to a fluence of approximately 5 × 10~9/cm2.)

4 GENERAL COMMENTS CONCERNED WITH PV HEAT TREATMENTS FOR PLM

A practical point is that a non-optimum heat treatment may not cause enough mechanical property recovery to economically justify the operation; the costs must be balanced against the potential time gained for continued electricity production. If copper-rich precipitates are TENSI.E STRENGTH

YIELD STRESS _

I

120

I

/

60--

//

/

/,,']

/ /

I

" r 120

I"

/!

/

//

I

,11

j/

-.1,,,=~-

// I 1

i

2

i

3

i

4

•40

/

¢'

o

"

r

/

nAR,/" "' //

I

1

8O

,.%

I~1./

/

20- ~

-20

I

x

80--

:I:

I

)/" "/

~/

100--

4O-

I

I

-20

'

0 I

2

I

3

I 4

5

FLUENCE [/1019 cm "2 ]

Fig. 4. Absolute changes in tensile properties at 150°C. Relatively more advantage due to intermediate annealing is indicated at fluences -<3 x 10~9/cm2.

Annealing for plant life management I

~-- 120

YELD STRESS I I

i 100

TENSILE STRENGTH I I I I

I /

221

/

t /

120

..'"""IT-- [-'oo II 1-80

80 60

/ II

//I

-

'^R/I

IJ'ml-oo

I[IN / II / IL/"

2O

/

,,/,¢

Io

1/ IIil ll -'°

tH IIMml 20

0

-20

I

I

I

I

I

I

i

I

1

2

3

4

1

2

3

4

5

FLUENCE 1/10 TMem -2 l

Fig. 5. Absolute changes in tensile properties as a function of fast neutron fluence. The testing temperature was 300"C. The rate of re-embrittlement for the IAR condition appears to be practically the same as for the I one.

known to be instrumental in the embrittlement, as in the case for this generic type of steel, 2-5 then annealing to overage them and remove more copper from out of solid solution is indicated. Annealing is likely to remove or favourably modify some irradiation induced defects and can thus contribute to the overall mitigation of neutron irradiation embrittlement. However, secondary effects which may cause embrittlement should be allowed for, for example, the possible segregation of phosphorus to grain boundaries aided by point defects. Non-hardening grain boundary embrittlement could occur in some high phosphorus steels (e.g. - 0 . 0 1 0 w t % ) ; this indicates that hardness measurements alone may not suffice to determine the efficacy of an annealing treatment. Other complementary mechanical tests, including fracture mechanical Table 1. Charpy notch toughness DB2T shifts and changes in USE for the A533-B mock-up PV steel

Condition

DBqT

USE (J)

Cv 41 J

Cv 68 J

I (0.4) I (2"2)

33 58

39 74

+12 -28

I (3.8)

80

88

-34

IAR (3.9) IAR (4.5)

75 64

85 75

-21 -25

(oc)

(oc)

I irradiated; IAR, irradiated + annealed + re-irradiated; multiply (fluences) by 1019/cnl 2, E > 1 MeV.

ones, appear to be essential in order to fully characterize the extent of mechanical property response to irradiation and annealing treatments. Other PVs, especially those of Russian manufacture, the so-called W W E R 440s and 1000s, have a basically different alloy, e.g. 15Ch2MFA which has 2.5% chromium and 1% nickel and molybdenum as alloying elements; the irradiation temperature is also lower (about 270°C) compared to the 290°C in Western PWR. Depending also on factors such as neutron fluence, energy spectrum and actual chemical composition and initial heat tratment applied in manufacturing the PV, it is found generally that the lower the irradiation temperature the higher the embrittlement induced. Attention should also be given to possible indirect effects due to ageing mechanisms (time at temperature) which may also contribute in part towards either the loss of mechanical properties or causing less recovery of them than anticipated for a given annealing treatment. The use of relatively high temperatures (i.e. ->500°C) should generally be avoided since microstructural changes could occur and also excessive thermal expansion effects could create additional plantspecific problems.

5 CONCLUSIONS • Irradiation caused hardening and embrittlement phenomena in the mock-up A533-B type

222

•

•

•

•

Philip Tipping, Robin Cripps

PV steel containing 0 . 1 4 w t % copper. Increases in the hardness, tensile yield stress and strength properties were found. The Charpy toughness was also affected, leading to increases in the D B T T and decreases in USE. Annealing, in this case, w h e n approximately 50% of the target fluence had b e e n reached, caused i m p r o v e m e n t to the mechanical property changes c o m p a r e d to the n o n - a n n e a l e d condition at similar final fluence levels. A t e m p e r a t u r e of 460°C for 18 h was used for this particular experiment. The annealing option for an actual P L M strategy should be i m p l e m e n t e d however, well before 50% of the foreseen E O L fleunce is reached. L o n g e r times than the one used here, for example up to 168h, do not appear counter-indicated if extrapolation of the time axis for the isotherms is p e r f o r m e d . A n n e a l i n g t e m p e r a t u r e s below 440°C would require relatively longer times for the recovery of hardness and other mechanical properties whilst t e m p e r a t u r e s above 500°C are not practicable and m a y even cause material degradation in their own right. The rate of re-hardening and e m b r i t t l e m e n t after irradiating, annealing and re-irradiating

(IAR-condition) was no faster than the e m b r i t t l e m e n t rate noted for the I condition. This indicates that the e m b r i t t l e m e n t mechanisms could be similar after this annealing treatment.

REFERENCES 1. Server, W. L., In-place thermal annealing of nuclear reactor pressure vessels. Report: NUREG/CR4212, EGG-MS-6708, 1984. 2. Hawthorne, J. R., Significance of copper, phosphorus and sulphur content to radiation sensitivity and postirradiation heat treatment of A 302-B steel. NUREG/CR0327, NRL Report 8264, 1979, April 12. 3. Odette, G. R., On the dominant mechanism of irradiation embrittlement of reactor pressure vessel steels. Scripta Metallurgica, 17(10) (1988) 1183-88. 4. Hawthorne, J. R., Koziol, J. J. & Byrne, S. T., Evaluation of commerical production A533-B steel plates and weld deposits with extra-low copper content for radiation resistance. In Effects of Radiation on Structural Materials, ed. J. A. Spague & D. Kramer. ASTM STP 683. American Society for Testing and Materials, Philadelphia, PA, 1979, pp. 235-51. 5. Tipping, Ph. & Waeber, W., Irradiation and annealing effects on the mechanical properties of a RPV steel and associated PLEX considerations. In Transactions of the lOth International Conference on Structural Mechanics in Reactor Technology (SMiRT), Vol. G, Los Angeles, 1989, pp. 239-44.