Effect of radial hydrides on the axial and hoop mechanical properties of Zircaloy-4 cladding

(1)

Eﬀect of radial hydrides on the axial and hoop

mechanical properties of Zircaloy-4 cladding

H.C. Chu

a,b,*

, S.K. Wu

a

, K.F. Chien

b

, R.C. Kuo

b a

Department of Materials Science and Engineering, National Taiwan University, Taipei 10617, Taiwan, ROC

b

Institute of Nuclear Energy Research, P.O. Box 3-14, Longtan, Taoyuan 32546, Taiwan, ROC Received 27 June 2006; accepted 13 November 2006

Abstract

The effect of radial hydrides on the mechanical properties of stress-relief annealed Zircaloy-4 cladding was studied. Specimens were firstly hydrided to different target hydrogen levels between 100 and 600 wt ppm and then thermally cycled in an autoclave under a constant hoop stress to form radial hydrides by a hydride reorientation process. The effect of radial hydrides on the axial properties of the cladding was insignificant. On the other hand, the cladding ductility measurements decreased as its radial hydride content increased when the specimen was tested in plane strain tension. A reference hydro-gen concentration for radial hydrides in the cladding was defined for assessing the fuel cladding integrity based on a cri-terion of the tensile strength 600 MPa. The reference hydrogen concentration increased with the specimen (bulk) hydrogen concentration to a maximum of90 wt ppm at the bulk concentration 300 wt ppm H and then decreased towards higher concentrations.

PACS: 62.20.Fe

1. Introduction

The mechanical properties of Zircaloy fuel clad-ding can be adversely aﬀected by the presence of hydrides, especially when they are oriented towards the radial direction of the tubing (i.e. radial hydride). Marshall and Louthan, Jr., demonstrated that Zircaloy-2 tube containing radial hydrides of

more than 50 ppm hydrogen exhibited no

macro-scopic ductility[1], whereas specimens with

circum-ferentially oriented hydrides at the same hydrogen

levels showed better ductility[2,3]. In order to retain

suﬃcient ductility to keep its integrity during reac-tor service, Zircaloy fuel cladding tube is manufac-tured to ensure that only circumferential hydride platelets are developed due to the hydrogen pickup from the waterside corrosion reaction. However, radial hydrides can be formed when a specimen is cooled down under stress from temperatures at

which a fraction of hydrides is dissolved [4–6]. As

doi:10.1016/j.jnucmat.2006.11.008

* _{Corresponding author. Address: Department of Materials}

Science and Engineering, National Taiwan University, Taipei 10617, Taiwan, ROC. Tel.: +886 3 4711400x6694; fax: +886 3 4711409.

E-mail address:[email protected](H.C. Chu).

(2)

higher burnups, the stress reorientation of hydrides is very likely to happen under some conditions

dur-ing spent fuel dry storage or reactor operation[7–9].

To assure the cladding integrity, Interim Staff Guidance-11, Revision 3 (ISG-11) is used by the US NRC staff when reviewing analyses of the poten-tial for spent fuel reconfiguration during storage conditions. It contains some limitations on the peak cladding temperature, cladding hoop stress and

repeated thermal cycling [10]. These acceptance

criteria are applicable for all commercial spent fuel burnup levels less than 45 GW d/MTU, and are proposed on the basis of the reduction in cladding ductility associated with the formation of radial hydrides. For spent fuel with higher burnups (exceeding 45 GW d/MTU), the analyses will be reviewed on a case-by-case basis because the current technical information is still insufficient. Therefore, a proper understanding of the mechanism responsi-ble for the stress reorientation of hydrides in the high-burnup fuel cladding is helpful to license appli-cation for spent fuel dry storage and transportation. However, the studies of radial hydrides and relevant influences on cladding tube with higher hydrogen levels are limited. In this work, the hydride reorien-tation behavior and its effects on mechanical prop-erties of the Zircaloy-4 cladding with hydrogen contents up to 600 wt ppm were investigated. Spec-imens were firstly hydrided to different target hydro-gen levels and followed by thermal cycling under a constant hoop stress to form radial hydrides. Then these specimens with a mixture of circumferential and radial hydrides were tested at room tempera-ture. The effect of radial hydride on the mechanical properties is discussed in this paper.

2. Experimental

2.1. Material and hydriding process

Stress-relief annealed (SRA) Zircaloy-4 cladding with an outside diameter of 9.5 mm and wall thick-ness of 0.58 mm was used in this study. Its chemical

composition is given inTable 1. Cladding tube, cut

into 13-cm lengths, was ﬁrst uniformly

hydrogen-charged by a thermal cycling process. The specimen was encapsulated with a pre-determined amount of pure hydrogen in a Pyrex capsule of suﬃcient vol-ume such that a low hydrogen partial pressure could be obtained to avert the formation of hydride layers. The encapsulated cladding specimen was then

ther-mally cycled between 200 C and 300 C for a

certain number of cycles, depending on the target hydrogen concentration level. The heating and

cool-ing rates were at 3C/min and 2 C/min,

respec-tively. The target hydrogen levels ranged from 100 to 600 wt ppm. Typically, hydrides were oriented in the circumferential direction and homogeneously distributed across the cross-section of the cladding specimen.

2.2. Hydride reorientation experiment

In order to obtain radial hydrides, the as-hydrided specimen was further subjected to thermal cycling in an autoclave under a constant hoop stress by regu-lating the diﬀerential pressure between its internal and external pressures with a constant diﬀerential pressure control system, as schematically shown in

Fig. 1. Prior to the hydride reorientation run, the

autoclave was evacuated and then ﬁlled with helium

gas of2 MPa and the cladding specimen was

inter-nally pressurized with water at room temperature.

Then the tube was heated at a rate of 3C/min to

400C under a constant diﬀerential pressure of

20.7 MPa that was equivalent to a hoop stress of

Table 1

Chemical compositions of Zircaloy-4 cladding tube (weight %)

Sn Fe Cr O N C H Zr

1.26 0.22 0.12 0.13 0.0029 0.01 0.0007 Balance

Dry air

Constant differential pressure control system

Autoclave Heater

Water

Booster & servovalve

Specimen

Fig. 1. Schematic diagram of the constant diﬀerential pressure control system.

(3)

160 MPa being applied on the tubing wall. After

solution annealed at 400C for 2 h, the specimen

was slowly cooled down at a cooling rate of 1C/

min to 170C to make up one thermal cycle.

Another thermal cycle started once the tube was

cooled down to 170C. Upon completion of the

thermal cycle treatment, the cladding tube was

furnace cooled from 170C to room temperature.

The pressure fluctuations due to thermal expansion of water and helium gas were regulated and mini-mized by the constant differential pressure control-ler, the maximum variation in differential pressure was less than 0.1 MPa. During thermal cycling, a fraction of hydride precipitates dissolved at higher temperatures. With the aid of the hoop stress, zirco-nium hydrides would re-precipitate out with their precipitate planes oriented in the radial direction of Zirclaoy cladding, when the specimen was cooled down. In this work, cladding tubes were treated under the same thermal parameter and differential pressure but different cycling numbers, i.e. 1, 2, 4, 8 and 12 cycles, to obtain specimens with various frac-tions of radial hydride precipitates.

Transverse sections of tubing specimens before and after thermal cycling were examined by optical microscopy to reveal the hydride morphology and orientation. The etchant used for metallographic

examination was composed of HF, HNO3, H2SO4,

and H2O in a volume ratio of 1:10:10:10. Hydrogen

concentrations of Zircaloy-4 cladding specimens were determined by an inert-gas fusion method using a LECO RH-404 hydrogen determinator. The sample for optical metallographic examination was cut from the same piece for hydrogen analysis. 2.3. Mechanical test

Following the reorientation process, the cladding tubes were subsequently cut in two pieces and machined into mechanical test specimens. Two types of test conﬁguration were used: uniaxial sion test (UTT) for axial loading and slotted arc ten-sion (SAT) test for circumferential loading. Detailed dimensions of these two specimens are given in

Fig. 2.

2.3.1. Uniaxial tension test

UTT tests were conducted on an Instron model 5582 mechanical testing machine at a nominal strain

rate of 1 · 10 4

s 1. A gripping device was

designed to provide suitable mate surfaces for the specimen curvature and lateral support to minimize

specimen bending that would result from the speci-men curvature and ﬁllet arc. The specispeci-men elonga-tion was measured by an LVDT extensometer over the mid-gauge section of 16 mm. The accuracy of the LVDT extensometer is ±0.2% of reading. 2.3.2. Slotted arc tension test

To simulate the loading conditions prevailing on fuel cladding in service, the SAT test was developed to determine the mechanical properties of Zircaloy-4 tubing material. Specimens were also tested on Instron mechanical testing machine at a nominal

strain rate of1 · 10 4

s 1. An anti-bending

mech-anism was added to the specimen grip with trough-shaped guides positioned on opposite ends of the

slot, as depicted in Fig. 3. A dual-head optical

extensometer was employed to take the strain mea-surements of an SAT specimen from both sides of its gauge section concurrently during testing. Then the strains were averaged to nullify the counter-acting bending eﬀects on both sides of the specimen section. The strain measurements taken by the opti-cal extensometer were veriﬁed and opti-calibrated by the strain gauge. The resolution of the extensometer is about 4 lm.

2.4. Analysis of hydride orientation

Since hydride platelets were inclined to precipi-tate in the form of long stringers and always linked together, the general orientation of hydride stringers

55.0mm 20.0mm 3.0mm 6.6mm 8.34mm 9.5mm 5mm 10mm 0.3mm 1mm

Fig. 2. Dimensions of (a) uniaxial tension specimen and (b) slotted arc tension specimen.

(4)

was thus selected to assess the hydride orientation rather than the specific orientation of individual platelets. The majority of hydride traces observed was either along the specimen hoop direction or per-pendicular to it, so it was convenient to classify the hydride stringers into two groups: circumferential and radial hydrides. The former was defined as the clusters with their precipitate planes oriented within 0–40 to the reference (circumferential) axis; the clusters within 50–90 to the reference axis were rec-ognized as radial hydrides. The small clusters within 40–50 to the reference axis were classified into neither of the two groups, and not counted into the total amount of hydrides.

The percentage of radial hydrides was deter-mined by calculating the areal fraction of radial strings on a photomicrograph. To provide a high-resolution digital image for this analysis, a digital

camera with a CCD array of 2048· 2048 pixels

was mounted on a microscope to project the hydride traces onto a screen. A magniﬁcation of 200 was selected. The orientation of hydride trace was recorded as each pixel in the image was scanned, the fraction of total pixel of hydrides in each cate-gory was then determined. On a photomicrograph, the hydride orientation analysis was conducted at two locations in the middle of the cladding wall. From the measurements, the average fraction of radial hydrides was then calculated. Because both

radial and circumferential hydride precipitate planes were predominantly parallel to the axial direction of cladding tube, all hydride reorientation data were measured on the transverse cross-section.

3. Results and discussion 3.1. Eﬀect of thermal cycling on hydride reorientation

An example of the reorientation of hydride pre-cipitates in Zircaloy fuel cladding is given in Optical extensometer

Grip

Anti-bending mechanism SAT specimen

Fig. 3. A sketch of the arrangement of a tension test on a slotted arc tension specimen.

Fig. 4. Micrographs showing reorientation of hydrides in Zirca-loy-4 cladding: (a) as-hydrided, (b) after 8 cycles of thermal treatment (230 wt ppm).

(5)

Fig. 4. Most hydrides were circumferentially aligned and uniformly distributed across the as-hydrided cladding wall. The majority of the hydride traces was thick and long; there existed a minor amount of fine hydrides. In general, both the coarse and fine hydrides had the same orientation. With increasing the number of thermal cycles, the proportion of these fine hydrides decreased and the reorientation of hydrides from circumferential to radial direction became more noticeable. The slow cooling rate of

1C/min provided suﬃcient time for hydrogen

atoms to diﬀuse and precipitate at their preferable sites.

The eﬀect of the thermal cycle number on the reorientation of hydrides in cladding tube is shown

to vary with hydrogen concentration in Fig. 5. A

hoop stress of 160 MPa was applied on the cladding tube while thermal cycling was proceeding. The per-centage of radial hydrides increased as the number of thermal cycles increased, until it reached a plateau

value on the reorientation curves plotted inFig. 5.

More than 90% of hydride precipitates in the 200– 300 wt ppm H specimens were reoriented into radial hydrides. For the specimen with a higher hydrogen content of 600 wt ppm, the applied stress of 160 MPa induced a maximum of about 20% radial hydrides after twelve cycles. Besides, a lower plateau value of approximately 78% radial hydrides was obtained for the 130 wt ppm H specimen. The fact that the extent of reorientation of hydrides in the cladding with 130 wt ppm hydrogen is lower than those of higher hydrogen content levels is believed to be related to the temperatures at which the

hydrides begin to nucleate from the saturated matrix.

According to the solubility data [11], a cladding

material with hydrogen contents higher than 200 wt ppm would start to have hydrides precipi-tated when it was slowly cooled from the holding

temperature of 400C, whereas precipitation in a

130 wt ppm H specimen did not occur until it was

cooled to 357C. The diﬀusion rate of hydrogen

atoms was smaller and the eﬀect of the stress on the hydride reorientation was less signiﬁcant in extent

at lower temperatures, relative to those at 400C.

Consequently, a lower plateau fraction value for radial hydrides was obtained on the cladding speci-mens with lower bulk hydrogen concentrations.

It is generally believed that stress reorientation takes place only on the hydrides which have dis-solved and then re-precipitated under stress. Hence small proportions of hydrides in the 320 and 600 wt ppm H specimens aligned radially after the ﬁrst cycle of heat treatment. A complete reorienta-tion of hydrides was attainable on the 320 wt ppm H specimen after 12 cycles of thermal treatment, whereas there were still about 120 wt ppm of

hydrides not dissolved at 400C during each cycle.

Results obtained in this work imply that, under a proper combination of cladding temperature and hoop stress, a complete reorientation of all hydrides is possible with the aid of repeated heating and cool-ing even though hydrides are not fully dissolved in each thermal cycle.

3.2. Hydrided Zircaloy cladding under tension tests

3.2.1. General description

The uniaxial tension specimens were loaded in a plane stress state and the slotted arc tension speci-mens in a stress state approaching the plane strain loading condition. A comparison of the typical stress–strain curves for UTT and SAT tests on Zircaloy-4 cladding specimens was exempliﬁed in

Fig. 6. Except for the specimens containing radial

hydrides, the SAT specimens sustain higher ﬂow stress than the UTT specimens but much smaller ductility, which is similar to the observations of

other researches on Zircaloy-2 [12,13]. The higher

ﬂow stress of SAT specimens is mostly attributed to the geometry of the plane strain tension specimen

[14].

Fig. 7shows the eﬀect of hydrogen concentration

on the mechanical properties of Zircaloy-4 fuel cladding tube tested under uniaxial and slotted

0 4 8 12 16 Number of cycles 0 20 40 60 80 100 Radial hydride (%) 250 ppmH 130 ppmH 600 ppmH 320 ppmH

Fig. 5. Eﬀect of the thermal cycle number on the hydride reorientation of cladding tubes with various hydrogen concentrations.

(6)

arc tension at room temperature. All hydrides in this case were homogeneously distributed and com-pletely oriented along the circumferential direction. The tensile strengths of both UTT and SAT fuel cladding specimens increased slightly with hydrogen content, their ductility values decreased as hydrogen concentration increased. These trends are in agree-ment with the observations by other investigators

[2,3,15,16].

Some of the as-received tubes were treated under the same conditions as those for stress-reorientation experiments and tested at room temperature to ver-ify the eﬀect of thermal cycle itself on the deforma-tion behavior of the material. Results of both UTT

and SAT specimens showed little or no dependence on the number of thermal cycles.

3.2.2. Eﬀect of radial hydride on uniaxial tension properties

The eﬀects of radial hydrides on the uniaxial ten-sile properties of Zircaloy-4 cladding specimens, with hydrogen concentration levels ranging between 130 and 600 wt ppm, tested at room temperature are

plotted inFig. 8. The eﬀect of radial hydrides on the

axial ductility of the cladding tube was insigniﬁcant even with the case of 320 wt ppm specimens in which most of hydride platelets were reoriented into

radial direction. This phenomenon could be

ascribed to the fact that both the face normals of radial and circumferential hydride platelets were perpendicular to the applied stress. The eﬀects of both circumferential and radial hydrides on the mechanical properties of cladding tube along the loading direction were similar.

On the other hand, when cladding tubes were subjected to tensile hoop stress, the radial hydrides with their platelet normals parallel to the stress direction and were susceptible to cracking along the hydride planes, a great loss in the circumferen-tial ductility was expected. The eﬀects of radial hydrides on the cladding hoop properties are dis-cussed in the following section.

3.2.3. Eﬀect of radial hydride on hoop tension properties

Fig. 9shows the eﬀects of radial hydrides on the

hoop tensile properties of Zircaloy-4 cladding

0 0.05 0.1 0.15 0.2 0.25 0.3 Strain (mm/mm) 0 200 400 600 800 1000 1200 Stress (MPa) SAT (0 ppmH) SAT (600 ppmH, 100% circumferential) UTT (600 ppmH, 100% circumferential) UTT (0 ppmH) SAT (130 ppmH, 78% radial) SAT* (600 ppmH, 20% radial)

* Fractured out of gauge section

Fig. 6. Typical engineering stress–strain curves for uniaxial and slotted arc tension tests on Zircaloy-4 cladding specimens at room temperature. 0 200 400 600 800 Hydrogen concentration (ppm) 0 200 400 600 800 1000 1200

Tensile strength (MPa)

0 10 20 30 40 50 Total elongation (%)

SAT, Tensile strength

SAT, Total elongation

UTT, Tensile strength

UTT, Total elongation

Fig. 7. Eﬀect of hydrogen concentration on the mechanical properties of SRA fuel cladding specimens tested under uniaxial tension and slotted arc tension; all hydrides in specimens were circumferentially aligned. 0 20 40 60 80 100 Radial hydride (%) 400 500 600 700 800 900

Total hydrogen content: 130 ppm, 250 ppm, 320 ppm, 600 ppm

Tensile strength

Total elongation Tested in uniaxial tension

Fig. 8. Eﬀect of radial hydrides on the mechanical properties of SRA fuel cladding specimens tested under uniaxial tension at room temperature.

(7)

specimens, with hydrogen concentration levels rang-ing between 130 and 600 wt ppm, tested at room temperature. It was observed that if all hydrides in the cladding were 100% circumferentially oriented, the test results showed a good reproducibility. Once some of the circumferential hydrides transformed into radial, the hoop mechanical properties of clad-ding deteriorated and the data became scattered.

Fig. 9 also suggests that some specimens with

signiﬁcant amounts of radial hydrides apparently have suﬃcient ductility, but that they were very brittle and fractured at stresses lower than the yield

strength. For example, the specimens with

600 wt ppm hydrogen failed when deformed in the elastic elongation range, but their stress–strain curves still showed nonlinear responses as if they

underwent plastic deformation (Fig. 6). The

contra-dictory observation could be ascribed to the forma-tion of some small surface cracks from breaking of radial hydrides during testing. Because the strains were taken by measuring the distance between the two indentations on the gauge section (0.7 mm), the formation of surface cracks caused an increment in this distance and thus higher strains were

obtained. Choubey and Puls[17]have used acoustic

emission (AE) to detect cracking of long radial hydrides in Zr–2.5Nb. They reported that cracking of hydrides was initiated in the low plastic region or slightly below the yield stress. And the small numbers of AE events generated in the early stage of deformation were not considered representative of hydride cracking because of unknown and uncontrolled stresses that might exist in the speci-men. In this study, surface cracks appeared

succes-0 20 40 60 80 100 Radial hydride (%) 0 20 40 60 80 100 Radial hydride (%) 0 20 40 60 80 100 Radial hydride (%) 0 20 40 60 80 100 Radial hydride (%) 0 200 400 600 800 1000 1200

Plane strain condition Tensile strength Total elongation Total hydrogen content: 130 ppm

0 200 400 600 800 1000 1200

0 200 400 600 800 1000

* * * * * * * * * * * ** * * * * *

* Fractured out of gauge section

Fig. 9. Eﬀect of radial hydrides on the mechanical properties of cladding specimens with various hydrogen content levels tested under slotted arc tension at room temperature: (a) 130 wt ppm, (b) 250 wt ppm, (c) 320 wt ppm, and (d) 600 wt ppm.

(8)

sively when specimens were gradually loaded up to 300 MPa. Because of the slight bending eﬀect dur-ing the initial stage of the test, most of the cracks tended to occur on the specimen inner surface to compensate the slightly diﬀerent stress levels on both sides of the slotted area.

Fig. 10 illustrates the fractographic features

of the specimens tested at room temperature. Normal dimples, ridges and round voids were the dominant features of the as-received specimens

(Fig. 10(a)), the brittle features of microcracks,

cleavages increased as hydrogen concentration

increased (Fig. 10(b)). For the specimens having

100% circumferential hydrides, the number of microcracks on the fracture surface increased with increasing hydrogen concentration. As can be seen

in Fig. 10(b) and (c) for specimens with the same

hydrogen content levels, the number of secondary cracks decreased if some hydrides reoriented into radial direction.

Fig. 10. SEM fractographs of SRA cladding specimens tested under slotted arc tension at room temperature: (a) SRA Zircaloy-4 with 7 wt ppm H (as-received), (b) 600 wt ppm H specimen with100% circumferential hydrides, and (c) 600 wt ppm H specimen with 21% radial hydrides.

(9)

To better understand the cladding deformation behavior along the hoop direction, toughness was used to evaluate the eﬀect of radial hydrides on the mechanical properties of cladding tube. The toughness was obtained by calculating the total area

under the stress–strain curve [18]. It was an

indica-tor which showed that the amount of work per unit volume could be done on a material prior to rup-ture. Unfortunately the concept of toughness could not give a clear trend of SAT test results. Since SAT test results were aﬀected unpredictably by a tiny var-iation of hydride distribution along the cracking path, the worst-case data of each test batch were selected to assess the eﬀect of radial hydride on the hoop mechanical properties of Zircaloy-4 clad-ding conservatively. The SAT test results were

re-plotted in Fig. 11. The concentration of radial

hydride inFig. 11was obtained by multiplying the

percentage of radial hydride and the bulk hydrogen content of the specimen. A residual strain of 0.01 was commonly used as an acceptance criterion in evaluating the integrity of fuel cladding, so 1% total elongation was taken as a reference value in

Fig. 11(a). It was found that cladding specimens

failed to meet this criterion when radial hydride

concentrations (reference concentrations) were

higher than 74, 157, 106, and 37 wt ppm for the specimens with bulk hydrogen contents of 130, 250, 320 and 600 wt ppm, respectively. Because of the fact that the surface crack probably occurred during SAT testing, the reference concentration of radial hydrides determined by cladding ductility

might not be conservative. As shown inFig. 6, the

yield strength of an intact Zircaloy-4 specimen (i.e. without surface crack) under plane strain condition

was 600 MPa. For this reason, a tensile strength

level of 600 MPa was chosen as an alternative acceptance criterion to determine the reference con-centrations conservatively. The reference radial hydride concentrations determined by 600 MPa for the specimen at each hydrogen level were 60, 100,

75, and 33 wt ppm, respectively (Fig. 11(b)). The

results are summarized in Table 2. The reference

radial hydride concentrations obtained in this work are comparable to those reported by Marshall and Louthan on the annealed Zircaloy-2 specimens with

total hydrogen less than 200 wt ppm [1,4]. They

suggested that all specimens with radial hydrides containing more than 50 wt ppm H exhibited no macroscopic ductility. It should be noted that the engineering strain was less than 1% when the yield point was reached during a ‘normal’ SAT testing (a test with no surface crack occurring). The refer-ence radial hydride concentrations determined by the cladding tensile strength were more conservative and reliable than those by the cladding ductility.

0 40 80 120 160 200 240 280 320 Radial hydride (ppm) 0 1 2 3 4 5 Total elongation (%)

Specimen hydrogen content 130 ppmH 250 ppmH 320 ppmH 600 ppmH 37 74 106 157

Tested under plane strain tension

0 40 80 120 160 200 240 280 320 Radial hydride (ppm) 0 200 400 600 800 1000 1200

Specimen hydrogen content 130 ppmH 250 ppmH 320 ppmH 600 ppmH 33 60 75 100

Fig. 11. Determination of the reference radial hydride concen-tration of Zircaloy-4 cladding material by the acceptance criteria of (a) 1% total strain and (b) tensile strength of 600 MPa.

Table 2

Reference concentrations of radial hydrides for brittle fracture of SRA Zircaloy-4 cladding with diﬀerent hydrogen content levels tested under slotted arc tension

Specimen Reference concentrations Determined by 1% strain Determined by 600 MPa stress (wt ppm) (%) (wt ppm) (%) 130 wt ppm H specimen 74 56.9 60 46.2 250 wt ppm H specimen 157 62.8 100 40.0 320 wt ppm H specimen 106 33.1 75 23.4 600 wt ppm H specimen 37 6.2 33 5.5

(10)

One factor that determines the mechanical prop-erties of cladding specimens is the continuity of hydride precipitates. As reported by Arsene et al.

[3], a ductile–brittle transition occurred when a

crit-ical inter-hydride spacing reached and the transition happened within a range of hydrogen contents (from 1500 to 2400 ppm). This work indicated, when parts of the circumferential hydrides became radial, the chance to form a continuous hydride net-work increased. However, the probability to reach the critical spacing between circumferential hydrides also decreased because some of hydrides were con-sumed as the reorientation process happened. It can be confirmed by comparing the hydride spac-ings in the specimens of the same hydrogen level but with different radial hydride contents. The con-tinuity of hydrides and the inter-hydride spacing were the two factors interacting with each other in a complicated way to affect the deformation behav-ior of Zircaloy cladding under slotted arc tension, which was reflected in wide variations of SAT test results from specimen to specimen.

From the data shown in the last column ofTable

2, the reference percentage of radial hydrides

line-arly decreased with increasing specimen hydrogen content. The reference hydrogen concentration increased with the specimen (bulk) hydrogen

con-centration to a maximum of90 wt ppm at the bulk

concentration 300 wt ppm H and then decreased

towards higher concentrations, as plotted in

Fig. 12. It could be accounted for by the hypothesis

that a large percent of radial hydrides are needed to develop continuous cracking path due to the large

inter-hydride spacings in the specimens of lower hydrogen contents, and that in the specimens of higher hydrogen contents, a reduction in hydride spacing would make it easier to link neighboring hydrides together with a demand for fewer radial hydrides to form a continuous network along the

cracking (Fig. 13). Besides, circumferential hydrides

would also work to exacerbate the brittle behavior when a cladding specimen had very high hydrogen contents. These results suggest that a small amount of radial hydrides can be extremely detrimental to

100 200 300 400 500 600

Specimen total hydrogen content (ppm)

0 50 100 150 200

Reference radial hydride (ppm)

SRA Zircaloy cladding Determined by 1% strain Determined by 600 MPa Predicted curve

Tested under plane strain condition

Fig. 12. Estimation of the reference radial hydride concentration as a function of specimen total hydrogen concentration.

Fig. 13. Cross-sections of Zircaloy-4 cladding specimens tested under slotted arc tension at room temperature: (a) 240 wt ppm H specimen with 98% radial hydrides, and (b) 620 wt ppm H specimen with21% radial hydrides.

(11)

the integrity of cladding materials of higher hydro-gen concentrations although they are diﬃcult to be formed.

4. Conclusions

In this study, the hydride reorientation behavior and the eﬀects of radial hydrides on the axial and hoop mechanical properties were investigated. Sum-marized below are the results:

1. Hydrided specimens with bulk hydrogen contents from 130 to 600 wt ppm were thermally cycled in an autoclave under a constant hoop stress. The percentage of radial hydrides increased as the number of thermal cycles increased until it reached a saturated value. More than 90% of hydride precipitates in the 200–300 wt ppm H specimens were reoriented into the radial direc-tion after several thermal cycles.

2. The hydride-reoriented cladding specimens with bulk hydrogen contents from 130 to 600 wt ppm were tested in uniaxial tension at room tempera-ture. The eﬀect of radial hydrides on the axial ductility of cladding tube can be neglected. 3. The eﬀects of radial hydrides on the hoop tensile

properties of Zircaloy-4 cladding specimens with hydrogen concentration levels up to 600 wt ppm were tested at room temperature using slotted arc tensile specimens. Test results were scattered but indicated a trend that mechanical properties degraded with increasing percentage of radial hydrides.

4. The reference concentration of radial hydrides for brittle fracture of cladding material increased as the total hydrogen content increased to about 300 wt ppm and then decreased with increasing

hydrogen concentration. The results in this study suggest that a small amount of radial hydrides can be extremely detrimental to the integrity of cladding materials of higher hydrogen concentra-tions although they are diﬃcult to be formed.

References

[1] R.P. Marshall, M.R. Louthan Jr., Trans. ASM 56 (1963) 693.

[2] J. Bai, C. Prioul, D. Francois, Metall. Mater. Trans. A 25 A (1994) 1185.

[3] S. Arsene, J.B. Bai, P. Bompard, Metall. Mater. Trans. A 34A (2003) 579.

[4] M.R. Louthan Jr., R.P. Marshall, J. Nucl. Mater. 9 (1963) 170.

[5] C.E. Ells, J. Nucl. Mater. 35 (1970) 306.

[6] M. Leger, A. Donner, Can. Metall. Q. 24 (1985) 235. [7] H.M. Chung, in: Proceedings of the International Meeting

on LWR Fuel Performance, Orlando, FL, 19–22 September 2004.

[8] G. Domizzi, G. Vigna, S. Bermu´dez, J. Ovejero-Garcı´a, J. Nucl. Mater. 275 (1999) 255.

[9] S. Shimada, E. Etoh, H. Hayashi, Y. Tukuta, J. Nucl. Mater. 327 (2004) 97.

[10] E.W. Brach, Cladding considerations for the transportation and storage of spent fuel, Interim Staﬀ Guidance-11, Revision 3, US NRC, Spent Fuel Project Oﬃce, 2003. [11] J.J. Kearns, J. Nucl. Mater. 22 (1967) 292.

[12] D. Lee, R.B. Adamson, in: A.L. Lowe Jr., G.W. Parry (Eds.), Zirconium in Nuclear Industry, ASTM STP-633, American Society for Testing and Materials, 1977, p. 385. [13] D. Lee, Trans. ASM 61 (1968) 742.

[14] D. Lee, W.A. Backofen, Trans. Met. Soc. AIME 236 (1966) 1077.

[15] R.C. Kuo et al., EPRI Report, TR-108753-P2, 2000. [16] J.-H. Huang, S.-P. Huang, J. Nucl. Mater. 208 (1994) 166. [17] R. Choubey, M.P. Puls, Metall. Mater. Trans. A 25A (1994)

993.

[18] G.E. Dieter, Mechanical Metallurgy, McGraw-Hill Book Company, London, 1988.