IAEA-SM-352/38

CREEP PROPERTIES OF NON-IRRADIATED ZrlNb CLADDING TUBES UNDER NORMAL AND ABNORMAL STORAGE CONDITIONS J. VESELY Skoda-UJP, Praha, a.s., Praha

XA9951806

M. VALACH UJV, Rez, a.s., Rez Z. FREJTICH, V. PRIMAN CEZ, a.s., Praha Czech Republic Presented by K. KLOC, Skoda-UJP, Praha, a.s. Abstract The Czech Republic has, for its long-term storage of WWER-440 spent fuel, opted for the dry storage alternative and uses the CASTOR 440/84 casks for this purpose. The maximum burnup of the spent fuel stored under a protective helium atmosphere in these casks is limited to 40 MW-d/kg U and the cladding temperature should not exceed 350°C. This temperature limit was established as a result of corrosion cracking analyses. However, with regard to the low pressure of the fission gases it can be anticipated that corrosion cracking will not be a critical mechanism of the cladding degradation. Thus, the damage is creep - controlled, as it was also indicated in the Russian paper [1] and the temperature of 350°C is unnecessarily conservative. This was a reason, why CEZ, a.s. (the Czech Power Utility), supported by the Ministry of Industry and Trade and the State Office for Nuclear Safety (SONS), initiated an experimental study of ZrlNb creep behaviour. This paper summarises experimental results of thermal creep of non-irradiated ZrlNb cladding tubes after recrystallization annealing, which modelled the creep behaviour of dry stored WWER spent fuel under normal and accident conditions. Determination of the storage limiting temperatures, at which there will be no creep caused damage over 40 and 70 year of storage period, taking into account the irradiation and ,,cooling" history of fuel, is also discussed. Zircaloy-4 specimens, exposed to creep similarly as ZrlNb ones, served to verify the tests methodology and the evaluation applied.

1. EXPERIMENTS 1.1. Tested materials and manufacturing of creep specimens Creep test specimens were made of standard ZrlNb and zircaloy-4 (Zry-4) cladding tubes supplied respectively by Russia and Sweden (SANDVIK). Their nominal external diameter (Do) and wall thickness (t) were: ZrlNb: Zircaloy-4:

D o = 9.178 ± 0.010 mm; D o = 10.178 ± 0.009 mm;

t = 0.709 ± 0.018 mm t = 0.712± 0.013 mm.

Zircaloy-4 tubes annealed to remove the cold forming stress have significantly higher strength properties. On the other hand, recrystallized annealed ZrlNb tubes have higher values of uniform elongation, both axial and circumferential. Mechanical properties of the tubes depend also on their oxygen content, see Table I. The full-length tubes of both alloys were cut into 120 and 100 mm sections (semi-products) which were then used to manufacture 200 ZrlNb creep test specimens and 140 Zry-4 specimens of the same kind. To determine the average ovality the wall thickness of each specimen was measured at roughly 25 perimeteral points at their half-length, and the external diameter - with an accuracy of 0.001 mm. The minimum wall thickness was marked on the specimens external surface - to establish a possible anomalous creep behaviour.

305

VESELY et al.

TABLE I. MECHANICAL PROPERTIES OF ZrlNb AND Zry-4 TUBES (AS SUPPLIED STATE) Alloy Tensile tests at 20°C :

Property [Unit]

ZrlNb XR

Zry-4 SR

Rmo [MPa] Ao [%]

223 407 39 105 198 50 461 568 45 280 36

550 720 21 300 412 25 745 896 12 480 12

[weight ppm]

500

1,000

Rpo.2 [MPa] Rm [MPa] A50 [%i

Tensile tests at 3 5 0°C :

Rpo.2 [MPa] Rm [MPa] A50 r%i

Burst tests at 20°C :

Rpo.2 [MPa] Rmo [MPa] A50 [%i

Burst tests at 3 5 0°C : Approximate Oxygen content

In the case of ZrlNb alloy, the initial dimension and their range are somewhat different from the nominal ones for the cladding tubes, mentioned above, this is immaterial. More important, however, is the fact that these tube specimens, as a result of the tube manufacturing technology applied, have a significantly larger ovality compared with that for Zircaloy-4 alloy. Differences between the maximum and minimum wall thickness are in the range of 15 to 80 mm. Such significant ovality indicates that a larger scatter of the results should be expected, as well as tendency to nonuniform deformation and its early localisation. Creep tests were carried out on the specimens with welded closing caps pressurised by pure argon: as a first step - both closing caps were electron-beam welded to the tube semi-products, one of these caps had a 0.7 mm filling-in hole. Actual pressurisation was carried out in a special evacuated and pressurized box equipped with a welding electrode. After a specimen was put into correct position, the box was several times evacuated and rinsed with argon and, finally, pressurized to the required pressure and then the hole was welded by using the TIG (tungsten inert gas) technique. 1.2. Testing equipment and its temperature characteristics 1.2.1. Testing equipment for the temperature range 325 - 400°C Creep tests were carried out in a laboratory muffle furnace provided with a proportional temperature controller and a maximum temperature cut-out. Two furnaces were operated at the same time, each for a different test temperature. Tests were performed in an air environment. The test specimens were placed horizontally into the heat-resistant tubes (block of tubes, each 150 mm long and 35 mm diameter) installed inside the furnace. 35 Creep specimens were tested for each selected temperature. The furnace temperature is measured with four jacketed thermocouples, fixed in the block of heat-resistant tubes. The measured temperature is periodically registered in the measuring computer unit. The maximum difference between data provided by the individual thermocouples was about 5°C, and the average furnace temperature was maintained with an accuracy of ± 4°C. The procedure of experimental results evaluation took into account the registered temperature differences. 1.2.2. Testing equipment for the temperature range 420 - 530°C To limit specimen oxidation, these tests were performed in an argon environment.

306

IAEA-SM-352/38

The equipment consisted of a resistance furnace with programmable temperature controller and instrumented tube retorts, made of sintered corundum. The retorts were equipped with valves, which enables to evacuate and fill them with argon, and with a cap, housing the draw rod of the specimens holder as well as four measuring thermocouples. The furnace volume was evacuated with a rotary vacuum pump, the residual pressure of < 10 Pa is sufficient. Argon was fed (up to atmospheric pressure] through the rubber bellows which during heating-up and cooling-down periods also served as a pressuriser. The whole capacity was 30 specimens, which were positioned horizontally and supported at both ends. The measuring ends of thermocouples were fixed within the specimen holder, in its upper and bottom parts, which enables to register the vertical differences for each experimental temperature. The pre-experimental temperature distribution and continuous temperature measurements, registered each 5 minutes for all thermocouples in the course of the actual creep tests, have provided the following characteristics of the test facility under stabilised conditions: • • •

temperature differences in the constant temperature zone (250 mm): < 4°C; vertical temperature differences in both test sections : < 0.5°C; the average temperature differences in both test sections : < 2°C.

1.3. Test conditions and experimental procedure Creep specimens pressurized up to basic pressure to achieve the hoop stresses in the range from 50 to 110 MPa for the temperatures of 325 - 400°C and from 40 to 130 MPa for the temperatures of 420 - 530°C, were weighted at analytical balance. The accurate hoop stress values were determined from the mass of argon and internal dimension of each specimen. Five stress levels were used for each testing temperature, always for three ZrlNb specimens or two Zry-4 specimens. The initial minimum and maximum diameter of the specimens has been measured along the whole length, in prior established intervals, by a digital micrometer. The length of the specimens was measured by a digital slide gauge. The actual exposures of the specimens at the selected temperatures were achieved by putting these specimens into the pre-heated furnace under atmospheric air conditions. Test in argon were started by moving the specimen holder into the pre-heated furnace without losing the argon protecting atmosphere. The actual beginning of the exposition was established on the basis of the specimens heating rate tests. After the required time had elapsed, the specimens were either removed from the furnace or moved into its cold part. A control weighting of the specimens was performed after each exposure, as well as dimensional measurements, and for significantly deformed specimens - also the volume determination. Usually, the creep tests were finished at each temperature after the demonstrable steady-state creep had been achieved for all specimens with the lowest hoop stress. In all cases, the end of the tests was decided in accordance with the ZrlNb alloy creep behaviour. The total time of the tests at the individual temperatures decreased from 9,600 hours at the temperature of 325°C to 23 hours at 530°C. 2. DETERMINATION OF THE CREEP EQUATION CONSTANTS FOR NORMAL AND ABNORMAL STORAGE CONDITIONS The experimental results were expressed in the form of creep curves - time dependence of the hoop strain. Stationary creep rate, saturated transient strain and the transient strain course were evaluated for each individual creep curve. The obtained results confirmed certain differences in the ZrlNb and Zry-4 creep behaviour. The primary (transient) stage for ZrlNb alloy is practically always shorter and its saturated transient strain is low. At higher temperatures tertiary creep sets in ZrlNb 307

VESELY et al.

earlier. The general tendency, to a higher creep resistance of recrystallized material compared with the annealed one, has also been confirmed at low stresses. The creep equation derived in [2] was used to model the time dependence of the deformation. The overall hoop creep strain of a tube specimen, creep stressed up to the steady creep region, is composed of the transient and stationary deformation parts and is described by the following set of equations:

where: 8 - hoop strain gf - saturated transient strain £^ - steady state creep rate

e,= Bx(E IT) e x hlr)

(3)

D = K x e{L-T)

(4)

where: a - applied hoop stress [MPa], E - Young modulus of ZrlNb alloy (E = 1.121 105 - 64.4 T), [MPa] [3] T - temperature [°K] Q - apparent activation energy [J/mol] R - gas constant (8.314 J/mol) B, C, F, G, H, I, K, L and n - experimentally obtained constants. To obtain the constants in the equations, the following procedure has been applied: 1.

2. 3. 4.

In evaluating the stress dependence of the stationary creep rate for individual temperatures within the anticipated range of equations validity, the C values for the individual temperatures are derived; Using the average C value, the temperature dependence of the stationary creep rate is evaluated and the Q and B values are obtained; In evaluating the dependence of the saturated transient strain on the stationary creep rate and temperature, the F, G, H and I constants are obtained; In evaluating the time dependence of the transient strain, the n and D constants are obtained; for the temperature range between 325 and 400°C the D constant is temperature dependent, which may be expressed by using the L and K constants.

The C and Q constants for Zry-4 were determined to verify the test methodology and the evaluation procedure. The obtained values C = 2,300 and Q = 207,000 J/mol were in good compliance with the results included into [2]. Two groups of constants for both the temperature and stress ranges included in Table II, were determined for ZrlNb alloy. For the interval of temperatures between 400 - 420°C the calculation is assumed to be in accordance with both models; a higher value of deformation for the given stress and time will be used in further applications.

308

IAEA-SM-352/38

TABLE II. CREEP EQUATIONS CONSTANTS Temperature

325 - 400°C

420 - 530°C

Hoop stress C B Q F G H I D

50- HOMPa 2,400 1.3 102 185,000 J/mol -0.0782 59.8120 -0.002410 2.2720 K = 56140 L= -0.009547 0.5

40 - 130 MPa 3,900 8.718 106 256,000 J/mol - 0.0700 53.2300 - 0.002767 2.4650 45

n

0.6

More accurate values of the creep equations constants will result from the experiments in progress, with gradually increasing and decreasing the temperature within the given temperature range, taking into account the model of the deformation hardening, taken from [4 and 5]. Fig. 1 and 2 show experimental values of the hoop strain in comparison with the values computed using the equations above, for the temperatures of 350°C and 450°C, that is for the normal and abnormal storage conditions.

Applied hoop stress [MPa]:

500

1000

1500

2000

2500

exposition [h]

FIG. 1. Comparison between experimental measured creep strain values and predicted creep curves at temperature 381 °C

309

VESELY et al. 35 Applied hoop stress [MPa]: 30

25

O

39,9 :

•

62,1

D • A

83,5 i 99,1 : 119,6:

•

20 c "to "55 o.

15

10

5 -

200

400

600

800

1000

exposition [h]

FIG. 2. Comparison between experimental measured creep strain values and predicted creep curves at temperature 453 °C 3. METHOD OF EVALUATION OF MAXIMUM ALLOWABLE TEMPERATURES OF THE SPENT FUEL CLADDING In accordance with the proposed criterion of 1 % total hoop strain for long-term storage, which should eliminate that the stored fuel rods may lose their initial tightness [6], a computer code was developed which is capable to calculate the maximum allowable WWER fuel cladding temperatures during cooling within CASTOR 440/84 casks. The code enables to calculate the limiting cladding temperatures in dependence of the internal gas pressure in the spent fuel rods with a burnup up to 40 MWd/kg U [7]. Fig. 3 presents the code block diagram. The computer code was tuned by using the creep equation for Zry-4 alloy taken from [2]. Implementation of the ZrlNb creep equations is foreseen after the evaluation of the experiments with increasing and decreasing temperature is finalised. The computer code used to calculate the increase of the cladding creep deformation may also serve to assess the progress of vacuum drying after a cask is filled, using the criterion of maximum 0.1 % of hoop strain [8]. The same code is applicable for the assessment of casks emergency states (insufficient cooling) by using the criterion of maximum 10 % hoop strain [9].

310

IAEA-SM-352/38

To input internal pressure in the spent fuel rod at 25°C and initial temperature

To calculate cladding temperature using spent fuel coolig profile and hoop stress in the cladding wall at each time intervals (45 steps )

Decrease initial temperature and repeat calculation

To calculate total hoop strain sc from creep equation for storage time 40 or 70 years ( using strain-hardening rule )

4.

To compare calculated total hoop strain sc with allowable strain

if

The highest temperature (when sc< 0,01 ) = max.allowable temperature for inputed internal pressure in the spent fuel rod at 25°C

To determine graphic dependency of MAT on iner wall pressure in spent fuel rod at 25°C for fuel cooling time 5,6,7 and 10 years; dry storage time 40 or 70 years and selected fuel burnup. FIG. 3. Block diagram of the computer code to determine the maximum allowable temperature of fuel cladding

31:

VESELY et al.

REFERENCES [1]

[2] [3] [4]

[5] [6] [7] [8] [9]

312

I.M. KADERMETOV, Y.K. BIBILASHVILI, A.V. MEDVEDEV, Evaluation of the maximum allowable temperature of WWER-1000 spent fuel under dry storage conditons, Proc. Int. Symp. On Safety and Engineering Aspects of Spent Fuel Storage, Vienna, 1 0 - 1 4 October 1994, IAEA, STI/PUB/949, Vienna, (1995) 269,279. M. MAYUZUMI, T. ONCHI, Creep deformation of an unirradiated Zircaloy nuclear fuel cladding tube under dry storage conditions, J. Nucl. Materials, 171, (1990) 381,388. S.E.LEMEKHOV, Computer Code ASFAD: Status, recent developments and applications, Int. Top. Meeting on LWR Fuel. M. MAYUZUMI, T. ONCHI, The applicability of the strain - hardening rule to creep deformation of Zircaloy fuel cladding tube under dry storage condition, J. Nucl. Materials 7379, 178,(1991). G.E. LUCAS, R.M.N. PELLOUX, Some bservations on time - hardening and strain - hardening rules for creep in Zircaloy-2, Nuclear Technology, Nol. 53, (1981) 46. M. PEEHS et al., Experimentally based spent fuel dry storage performance criteria, Proc. Third Int. Conf. Spent Fuel Technology,Seattle, 1986, CONF-860-417, p. 316. J. ZYMAK, Calculation of spent fuel behavior during dry storage and determination of limit storage conditions, UJV Z-260-T,M, (1997). I.S. LEVY et al., Recommended temperature limits for zircaloy - clad LWR spent fuel dry storage in an inert atmosphere, PNL-6189, (1987). M. MAYUZUMI, T. ONCHI, A method to evaluate the maximum allowable temperature of spent fuel in dry storage during a postulated accident, Nuclear Technology, vol. 93, (1991) 381,388.