FUNDAMENTAL METALLURGICAL ASPECTS OF AXIAL SPLITTING

IN ZIRCALOY CLADDING*

by

H. M. ChungArgonne National Laboratory

Argonne, IL 60439, USA

‘fix mlbmiffed mamlmipf h been Creafrd by If), uidm+’ Of

c+ = CP=ator of Awmne Nati.naf LaLvrafoty ~ArgOnne~

under C-antract N.. W-31- 109-ENG-38 with fhe U.S. Department of

Enqy. ‘flw U.S. Cwanmmt retains for ifscX and others acting on

* tx.ML a paid-up, nomxck+% irrevocable wurldnide km= in

said atticfe to reproduct. prepare dcrivadvc uwks, distribwc

c+= fO ~ pub% and tiorm publkfy emd dkplq pubtidy, by

or 0“ W of fhc Gwanln.t.

February 2000

To be published in the Proceedings of the International Topical Meeting on Light WaterReactor Fuel Performance, April 10-13, 2000, Park City, Utah.

*Work supported by the U.S. Nuclear Regulatory Commission, Office of Nuclear RegulatoryResearch.

— ——. —._.

DISCLAIMER

This report was prepared as an account of work sponsoredby an agency of the United States Government. Neitherthe United States Government nor any agency thereof, norany of their employees, make any warranty, express orimplied, or assumes any legal liability or responsibility forthe accuracy, completeness, or usefulness of anyinformation, apparatus, product, or process disclosed, orrepresents that its use would not infringe privately ownedrights. Reference herein to any specific commercialproduct, process, or service by trade name, trademark,manufacturer, or otherwise does not necessarily constituteor imply its endorsement, recommendation, or favoring bythe United States Government or any agency thereof. Theviews and opinions of authors expressed herein do notnecessarily state or reflect those of the United StatesGovernment or any agency thereof.

,-,--- -r-z-v ., ,.. ,, .X,. ,.?. ., .,.-,- y-, -, . . .- -- —. —~.—, --= —“. .- . . . . . . . .-—

DISCLAIMER

Portions of this document may be iilegiblein ‘electronic image products. Images areproduced from the best available originaldocument.

FUNDAMENTAL METALLURGICAL ASPECTS OF AXIAL SPLITTING INZIRCALOY CLADDING*

H. M. ChungArgonne National Laboratory

Argonne, IL 60439, USAheechung(ijanl.gov

ABSTRACT

Fundamental metallurgical aspects of axialsplitting in irradiated Zirc~loy cladding havebeen investigated by microstructuralcharacterization and analytical modeling, withemphasis on application of the results tounderstand high-bumup fbel failure under RIAsituations. Optical microscopy, SEM, andTEM were conducted on BWR and PWR fhelcladding tubes that were imadiated to fluencelevels of 3.3 x 1021 n cm-2 to 5.9 x 1021 ncm-2 (E > 1 MeV) and tested in hot cell at292-325°C in Ar. The morphology,distribution, and habit planes of macroscopicand microscopic hydrides in as-irradiated andposttest cladding were determined by stereo-TEM. The type and magnitude of the residualstress produced in association with oxide-layergrowth and dense hydride precipitation, andseveral synergistic factors that stronglyinfluence axial-splitting behavior wereanalyzed. The results of the microstructuralcharacterizations and stress analyses were thencorrelated with axial-splitting behavior o fhigh-bumup PWR cladding reported forsimulated-RIA conditions. The effects of keytest procedures and their implications for theinterpretation of RIA test results are discussed.

INTRODUCTION

Long axial splitting of irradiated Zircaloycladding at <400°C has been observed underconditions such as continued operation ofdefective fiel rods in some boiling waterreactors (BWRS), pulse testing of high-bumupfuel in test reactors under reactivity-initiated-

*Work supported by the U.S. NuclearRegulatory Commission, Office of NuclearRegulatory Research.

-accident-like (RIA-like) conditions, and hot-cell testing of subcritical crack propagationand burst behavior of spent-fiel cladding. Theseverity of axial splitting (i.e., length, width,and number density) is an important factorthat influences potential release of fiel piecesor particles; therefore, it is important tounderstand the phenomenon from thestandpoint of not only fiel economy but alsoreactor safety considerations. Axial splittingof BWR fhel during normal operation [1-5],which usually occurs in defective cladding thatcontains a pinhole-type primary breach, hasbeen attributed to various mechanisms such asdelayed hydride cracking [1,3], corrosionhydride cracking [4], and hydrogen-assistedlocalized shear [2,4]. Results of extensiveinvestigations have also been reported foraxial cracking in CANDU reactor’ pressuretubes fabricated from Zr-2.5Nb or Zircaloy-2[6-12]; such cracking has been explained onthe basis of delayed hydride cracking. Axialsplitting under RIA-like situations has beenobserved in pressurized water reactor (PWR)cladding at very high strain rates, usuallywithin <100 ms of pulse loading at =20 or=290”C [13-18]. For this type of situation, itis obviously difficult to explain the fast axialsplitting on the basis of delayed or corrosionhydriding mechanism, simply because the

strain rate (on the order of =1 s-1) is too highto allow sufficient migration of hydrogenatoms toward the advancing crack tip. Failureof medium- [19] and high-burnup fiel cladding[20-22], investigated by hot-cell burst [19-22]or expanding-mandrel [19] testing at =292-325°C, was also via long axial splitting. Incontrast, failure of short ring specimens from=50-MWd/kgU PWR fuel cladding, testedunder uniaxial tension in the radial direction at=343°C, was not via axial splitting [23].However, some ring tension specimens, cutfrom =63-MWdlkgU PWR fuel cladding andtested at =280”C at the high strain rate of =5

/

—.-.— ----- ——2 ------. .._— . . ..~- -—-

. . . /

., . . I

S-l, exhibited low ductility and straight axialsplitting [15].

Apparently, long axial splitting occurs undervarious conditions of loading (thermalexpansion of pellets, internal gas pressure,uniaxial tension, or stress intensity), strainrate (=10-8 to 1 s-l), temperature (==20 to=400°C), fluence, hydrogen concentration,and hydriding. Fracture surface morphologiesalso vary, e.g., pseudocleavage, striations,chevron or pseudochevron pattern, steppedterrace, shear tear, ductile dimple, or acombination of these. Although thephenomenon has been investigatedextensively, the mechanism of in-reactorsplitting is still not well understood.Irradiation is apparently not a necessarycondition, because long axial splitting has beenobserved in nonirradiated hydrided specimens.Therefore, presence of hydrogen and hydridesappears to be the only microstructural aspectknown to be common to axial splitting undervarious conditions. Hydride reorientation nearthe crack tip appears to be a prerequisite insome situations (e.g., laboratory simulation ofdelayed-hydride crackh@ but not necessarilyin others (e.g., hot-cell burst test and pulse testunder RIA-like conditions). Therefore, in thiswork, fimdamental metallurgical aspects ofaxial splitting in irradiated cladding have beeninvestigated by microstructural analysis andanalytical modeling, with emphasis on possibleapplication of the results to better understandthe axial splitting behavior of high-burnupcladding under RIA situations. Opticalmicroscopy and scanning and transmissionelectron micropscopies (SEM and TEM) wereconducted BWR and PWR fuel cladding,irradiated to 3.3 x 1021 n cm-2 to 5.9 x 1021 ncm-2 (E > 1 MeV) and tested by inert-gaspressurization at 292-325°C, to gain a betterunderstanding of the micro structuralcharacteristics and residual stress associatedwith hydriding and oxide layer growth. Newfeatures of hydride-related microstructure ofirradiated Zircaloys were revealed by TEMthat were not visible by optical microscopy orSEM. Results of the microstructuralcharacterization and analytical modeling ofassociated residual stress were then correlatedwith the axial splitting behavior of irradiatedfuel cladding reported for RIA-like situations

I!

to better understand the influence of key testprocedures and to critically evaluate the .

relative failure behavior under conditions ofsimulated pulse test and in-reactor RIA.

EXPERIMENTAL PROCEDURES

BWR and PWR fuel cladding tubes, irradiatedto 3.3 x 1021 n cm-2 to 4.4 x 1021 n. cm-2 (E

.> 1 MeV), respectively, and tested by gaspressurization or expanding mandrel at 292-325°C, were selected for microstmcturalanalysis by TEM and SEM. Details of thestress-rupture and expanding-mandrel testshave been reported elsewhere [19]. Axialsplitting in the specimens occurred in 1-312 h.Diametral strain at failure ranged from 0.4 to11.7 ‘%o, which corresponded to averagediametral strain rates of =5.4 x 10-9 S-l to=3.0 X 1o-6 s-l. In addition to themechanically tested materials, TEM was alsoconducted on as-irradiated PWR fbel claddingirradiated to 5.9 x 1021 n cm-2 (E > 1 MeV).No mechanical testing was conducted on thelatter higher-fluence Zircaloy-4 cladding.

Table 1 summarizes the materials, irradiationhistory, and test conditions.

TEM specimens were prepared by the

procedure shown schematically in Fig. 1. Astress-ruptured or mandrel-cracked claddingtube was cut and split to reveal the fracturesurface along the axial split. Afler thefracture surface morphology was characterizedin an SEM, narrow strips, which contained thefracture surface in one side of the strip, wereprepared. The slightly curved strips wereground to =0.2-O.3-mm-thick flat strips.Then, several standard-size TEM disks werepunched out of the thin flat strips. Thus, theplane of the TEM thin-foil specimen wasnearly parallel to the tangential direction ofthe cladding midwall. The TEM specimen wasthen jet-polished in a solution that contained25 ml perchloric acid, 225 ml acetic acid, and50 ml butylcellosolve and was maintained atabout -70”C in a bath of dry ice and methanol.TEM analysis was performed at =23°C at 100keV in a JEOL 1OO-CXII scanningtransmission electron microscope or at 1000keV in a Kratos/AEI EM7 high-voltageelectron microscope.

I

Ac+

Table 1. Summary of material, irradiation history, and mechanical-testconditions of Zircaloy fuel cladding examined by TEM

Reactorand

Material

BWR, standard

Zircaloy-2

PWR, standard

Zircalov-4

PWR, standardZircalov-4

CladdingTube

ID

165AG1O

165AE4A

165AE4B

217A4G

600W8

Fluence Test Time to(1021 n cm-2, Temp.

DiametralLoading Failure Strain at

E >1 MeV) ~c) Method (h) Failure (Yo)

3.3 I 325 I Ar r)ressure I 1.0 I 1.1

3.3 I 325 I Ar uressure I 25.1 I 0.8

3.3 I 325 I Ar txessure I 207.6 I 0.4

4.4 325 mandrel expansion <335.2 3.6

5.9 - not tested(as-irradiated)

it

Axial Splitin Tube

Figure. 1. Schematic illustration of procedure used to prepareTEM specimens from irradiated fuel cladding.

OXIDE-INDUCED STRESS

The radial cracks observed in the internallypressurized BWR-irradiated Zircaloy-2specimens listed in Table 1 had always beeninitiated at the outer-diameter (OD) side ofthe cladding. Examples of such cracks,observed on the metallographic Specimen165AG 10 (Table 1) are shown in Fig. 2. Inthe cladding tube that failed in =1.0 h(diametral strain at failure, =1. l%), more than40 incipient radial cracks were observed alongthe circumference of the OD-side metal. Mostof the incipient cracks penetrated up to =200#.m toward the inner diameter (ID) of the walland then stopped when the crack tip reached

the ductile metal that contains a lowerconcentration of oxygen. Numerous crackswere also observed in the oxide; they fullypenetrated the oxide layer. Practically all ofthe penetrating cracks in the oxide wereassociated with at least one incipient crack inthe adjacent metal. As shown in Fig. 2,however, there were regions in which nocorresponding oxide crack was observed in thevicinity of a crack in the metal. This findingindicates that the OD-side metal covered withan oxide layer is more susceptible to crackingthan the oxide itself.

The oxide layer in some local regions of theBWR cladding could be separated from themetal without much difficulty. An example is

3

——— —.- .. ..-

-.. I

shown in Fig. 2. In the metallic region overwhich the oxide layer was removed, no or anegligible number of cracks were observed inthe OD metal. This observation indicates thatsignificant oxide-induced residual stress w=present on the oxide-covered metallic region.When the oxide was removed or spalled, theresidual tensile stress seemed to have beenrelieved; therefore, no cracks were produced inthat region. That is, oxide-induced residualstress seems to be a major factor in thecracking of the OD-side metal.

J

.

,

,I

I

—— ...-

Figure 2.Crack patterns in BWR Cladding 165AG 10 that was stress-mptured by internal Ar gaspressurization, showing absence of cracks in OD metal over which the oxide layer,and hence, residual tensile stress, was removed.

Because of the large volume expansionassociated with oxidation of Zr metal to Zr02(i.e., the Pilling-Bedworth ratio of Zr is aslarge as =1.56), the metallic layer in contactwith the waterside oxide layer will be subject toresidual tensile stress, whereas the oxide will besubject to compressive stress. Figure 2indicates that oxide-induced residual stressplays an important role in crack initiation ordeformation:discussed byimplicated thestress on axial

A similar effect has beenDonaldson [24], who hasinfluence of oxidation-inducedgrowth of unirradiated oxidized

Zircaloy-4. The oxidation-induced residualstress can be calculated to a firstapproximation by using the model shown inFig. 3. Results of such calculations aresummarized in Table 2. The magnitude of theresidual tensile stress and associated stressintensity on a flaw in the metallic layer is afunction of the thickness of the innermostadherent oxide layer. The typical thickness ofthe adherent innermost oxide ranges from 2 to4 pm for PWR cladding, although a thicknessas great as =10 ym has been observed in somecladding. The adherent innermost oxide layer(denoted as Lox in Fig. 3), which contains ahigh volume fraction of tetragonal Zr02,seems to be present always in PWR Zircaloy-4cladding regardless of bumup, fluence, or theoverall oxide structure. Another importantcharacteristic of the OD side metal is themultiaxial state of stress that is significantlymodified by the residual stress produced by the

large expansion in volume of Zr02. Thedegree of this modification will varysignificantly from region to region, especiallywhen burnup is high and when many shorttangential cracks are produced in the oxide, asin high-burnup standard Zircaloy-4 cladding.

Table 2. Estimated oxide-induced residualstress and stress intensity for flaws in metalliclayer in contact with OD oxide layer of PWRcladding (applied stress is assumed to be 138MPa)

For Lox = 3 pm

Flaw size I l~ml 10 pm

Residual stress (MPa) 200 200

Total stress (MPa) 338 338*

Stress intensity (MPa m“2) 0“4 ‘“3

ox-de vdum e

Sax, com Pr--veexpands

1.56 timesstress in cxide

4_f, ‘%4?::. - :+-.::-:’.”’--.-”;— -:?,.; . . ,,.,.-,.,,:>. _ ,. _,: ,. . .— ‘“

L Sm, tensile stress in metal

thickness ofadwrent wide

moxExHmLoxsw

SmHm W+

metalaideYoung’s modulusetastic strtinmetal layer thicknessadmrent wide thicknessotidation-induced stresscommon width of layer

Sox Lox W = O

Sm = Em M (tensile stress in metal)Em M Hm - Eox (0.56/3 - Xox) Lox = O

XCM = negligibleEox = ZOO GPaHm = 560 pm for PWR highburnup claddng

Sm (in Ma) = 66 %x (LOX in pm)

Figure 3.Summary of model designed to calculate oxide-Iayer-induced residual stress in cladding metal.

5

. . .. .. -,-. -.—— ——--~ -,, --—- -_. . —..—- —. ...?r--’-- T. ...

.

CHARACTERISTICS OF HYDRIDEPRECIPITATION

Macroscopic Hydrides

In an operating PWR, =12-18% of all Hatoms that are generated by the watersideoxidation are picked up by the cladding [20].When the volubility limit of H, i.e., =90-100wppm at =325°C in irradiated Zircaloy [25], isexceeded, hydride platelets precipitate. Athigh bumup, the metallic layer in contact withthe waterside oxide layer in PWR claddingusually contains a high density ofcircumferential hydrides, while the density ofhydrides in the midwall or in the metallic layerin contact with fuel pellets is low.

Macroscopic hydride platelets that aredistributed perpendicular to the direction ofprimary stress (i.e., radial hydrides) areparticularly deleterious to the cladding’smechanical properties. However, suchorientation is not typical; macroscopichydrides under normal operating conditionsprecipitate mainly as platelets in thecircumferential direction. Development ofradial hydrides, usually referred to as hydridereorientation, can occur under variousconditions. Yet, characteristics of hydridereorientation in high-burnup cladding, such asthe effects of applied stress, texture,temperature, habit plane, and cooling rate, arenot well understood.

Hydrides preferentially precipitate on acertain habit plane of the unit cell of the hcpa-phase Zircaloy metal. The structure, size,morphology, distribution, habit plane, andorientation of hydrides, and the residual stressthat is developed in association with hydrideprecipitation, play important roles in themechanical properties of irradiated claddingunder uniaxial or biaxial loading conditions[20-23]. These characteristics have beeninvestigated. extensively for unirradiatedhydrided cladding by optical microscopy andTEM. However, most investigations onhydride characteristics in irradiated fuelcladding have been limited to analysis byoptical microscopy; few have included themore powerfhl technique of TEM.

Habit Plane of Circumferential Hydrides

It has been generally agreed that the habitplane of hydride precipitation in unalloyed cz-phase Zr is {100)zr, i. e., the prism p17iii&[26-28]. In a surprising contrast, Westlake [29]reported that the habit plane of hydrides inunirradiated unstressed Zircaloy-2 and -4 is{107)Zr, a plane very different from the prismplane but rather similar to the basal plane.“The orientational relationship be~een the fcc5 hydride and the unalloyed hcp Zr has beenrepresented as [28]:

(11l)a-hydrideil(ool)a-zr[1 ~o]/j-hy&-i&j//[llo]a-zr

The recrystallized specimens of Westlake werecharged with hydrogen gas at =770°C with nostress applied on the specimens. Therefore,the material, stress state, and the hydridingconditions significantly differ from those offuel cladding in an operating LWR.Therefore, in this study, microstmcturalcharacteristics of hydrides were analyzed byTEM using actual spent fuel cladding.

Midwall regions of BWR and PWR claddings(Table 1) that contained hydrides wereanalyzed by optical microscopy and TEM. Atypical example of the hydride’ morphologyand orientation, as resolved by opticalmicroscopy, is shown in Fig. 4. These“macroscopic” hydrides in the tangential-radial and axial-radial planes in the figure areplatelets that are oriented nearly parallel tothe circumferential direction of the claddingtubq usually =2-10 pm wide and =0.1-0.2 ~mthick, they are readily revealed by opticalmicroscopy.

We characterized the “macroscopic” hydridesthat were in the electron-transparent regionsnear the cladding midwall by TEM; typicalexamples of bright- and dark-field images areshown in Fig. 5. As indicated by the directionof [107]Zr in the figure, the habit plane of thehydride is close to {107}zP All other“macroscopic” hydrides in other examinedregions exhibited a similar habit plane veryclose to {107}ZP Therefore, we conclude thatthe habit plane of macroscopic hydrides .

6

Figure 4.Optical images of macroscopic 5 hydrides: (left) PWR Cladding600W8 irradiated to 5.9 x 1021 n cm-2 (E > 1 MeV), tangential-radialplane, ID to midwall; (right) BWR Cladding 165AE4A irradiated to3.3 x 102i n cm-2 (E >1 MeV), axial-radial plane, near midwall.

in irradiated Zircaloy-2 and -4 fuel cladding is{107}Zr, the same as that reported forunirradiated unstressed Zircaloys by Westlake[29]. As shown in Fig. 5, the {107}Zr plane isonly 14.7° away from the basal plane, {001}Zror {002}ZP Therefore, the basal plane is ausefil reference plane for the habit plane ofhydride precipitation in Zircaloy cladding. Aresult of the crystallographic relationshipbetween the habit plane of hydrides and thetexture of commercial fuel cladding is thatmost hydrides precipitate in thecircumferential direction of the cladding.

Habit P1anc of Reoriented Hydride

A significantly bent hydride was observed in aPWR cladding that was stressed to failure at=325°C by mandrel expansion; see Fig. 5,bottom-riexhibitedplane {

;ht inset. The hydride, whicha habit plane close to the prismOo}Zr, appeared to have been

deformed under the influence of stress duringthe expanding-mandrel test. It seems thatduring the mandrel test at =325°C, some ofthe hydrides in the specimen dissolve& andwhen the failed specimen was cooled slowly toroom temperature, some hydrides precipitatedon the {10O}Zr planes and were subsequentlydeformed under the influence of stress. This islikely because the failed specimen was stressedcontinuously by the expanded mandrel, evenduring the cooling period. This observation,then, indicates that, in contrast to the {107}zrhabit plane of the normal circumferentialhydrides, the habit plane of reoriented radialhydrides is the prism plane. However, afurther investigation is needed to confirm thisobservation.

Hydride Deformation

Another important observation is that, asindicated by the morphology of the hydride inPWR Cladding 217A4G in Fig. 5, hydrides are

—.-.

I.

Figure 5.

TEiM images of macroscopic 3 hydrides: (top left) BWR Cladding 165AG 10, bright- anddark-field images; (top right) BWR Cladding 165AE4A, bright-field; (bottom left) BWRCladding 165AE4A, bright-field; (bottom right) PWR Cladding 217A4G, dark-field.

ductile at and near the operating temperature.In the present TEM analysis, broken hydrideswere observed in neither the BWR nor thePWR cladding. However, as shown in thebottom left inset of Fig. 5, dislocations wereobserved frequently near macroscopichydrides. These observations indicate thatrather than fracture of the hydride itself,plastic deformation that occurs at themetal/hydride boundaries is the key factor inhydride-related degradation (axial splitting) ofmechanical properties in irradiated Zircaloycladding.

Microscopic Hydrides

Another important result of the TEM analysiswas the observation of “microscopic” hydridesin irradiated fiel cladding, i.e., small hydridesthat are not resolved by optical microscopy,For example, the bright-field image in Fig. 6shows several “microscopic” hydrides in themidwall region in the BWR cladding that wasirradiated to =3.3 x 1021 n cm-2 (E > 1 MeV).These hydrides were =100-200 nm long and=30-50 nm wide, too small to be -resolved byoptical microscopy. The reflections of thehydrides were split in the directions of [002]Zrand [0 10]Zr, as shown in the selected-arediffraction pattern in Fig. 6.

hydrides inited to =3.3left) bright-

with letterdiffraction

lm c%Zr, 8!kide.

9

..,...—=-- —n—-.. .--T=’-7--’-..,.’,—,--’...“$‘.““—-------”y--,~~--.-.,-.~,,:

._. .-. -..———_ I

The pattern also contains reflections from ccZr and artifact surface hydrides [30]. Themicroscopic hydrides shown in Fig. 6 exhibitthe following orientational relationship:

(3 10)6-hydride//(100)u-Zr[002] 3-hyd~ide//[O02]~-Zr.

In PWR Cladding 600W8, irradiated to afluence of =5.9 x 1021 n cm-2 (E > 1 MeV),microscopic hydrides were observed in highernumber density (Fig. 7). The microscopichydrides in the material (=20-80 nm thick and=100-500 nm long, Fig. 7) were much smallerthan the macroscopic hydrides (=100-200 nmthick and =2,000- 10,000 nm long) shown inFigs. 4 and 5. To show that the microscopichydrides are indeed bulk hydrides rather than

artifact surface hydrides that are oftenproduced during jet polishing of a TEMspecimen, stereo pairs of bright- and dark-fieldmages were obtained by tilting the specimen.An example of such dark-field stereo pairs isshown in Fig. 7. The dark-field images wereproduced with a reflection that contains[oo2]&hydri&, [Ooz]zr, and [Ooz]sutiace-hydtide

[30]; therefore, many irradiation-induced

“defects and <c>-type dislocations are visible inthe dark-field images in addition to small &phase “microscopic” hydrides. Whenexamined under a stereoscope, the dark-fieldpair clearly shows that the microscopichydrides are indeed bulk hydrides and that theyare nucleated in association with short <c$-dislocations. The dark-field images in Fig. 7also reveal the distribution of the microscopic

Figure 7.

TEM images of microscopic6 hydrides in PWR Cladding 600W8irradiated to 5.9 x 1021 n cm-2 (E >1MeV): (left) bright-field image, tiltangle OO; (bottom left) dark-fieldstereopair, tilt angle OO; (bottomright) dark-field stereopair, tilt angle3°. Letters “H” and “M” denote,respectively, hydrides and partiallyamorphized Zr(Cr,Fe)2 intermetallics.

. I

tI

II

[

I

t I

I

$

hydrides more clearly. For the PWR Cladding

217A4G, irradiated to =4.4 x 102I n cm-2,microscopic hydrides were observed only inlimited regions. This finding indicates thatprecipitation of microscopic hydrides is morepronounced for higher fluence and higher Huptake. The number density of themicroscopic hydrides in the TEMphotomicrograph of Fig. 7 appears to be atleast a few orders of magnitude higher thanthat of the macroscopic hydride shown in theoptical photomicrograph of Fig. 4. Like thatof the BWR cladding (Fig. 6), the habit planeof the microscopic hydrides in the PWRcladding (Fig. 7) was basal plane, i.e., {002}zPClearly, irradiation-induced <c>-typedislocation loops and short line dislocationsare the preferential sites for nucleation ofmicroscopic hydrides.

When the structure of a circumferentialmacroscopic hydride, such as those revealed byoptical microscopy in Fig. 4, is examined inhigh magnification, one can often see that themacroscopic hydride actually consists of manytiny hydrides that are closely aligned in aslightly different orientation. Thisobservation indicates that microscopichydrides, such as those shown in Fig. 7, areprecursor hydrides which eventually alignthemselves closely, coalesce, and develop intoa macroscopic hydride during operating.Furthermore, the facts that the numberdensity increases with burnup drastically andthat they are predominantly associated with~> dislocations suggest that microscopichydrides are formed during medium- to high-burnup operation and play an important rolein the mechanical properties of fuel cladding,especially in the hydride-rich outer region ofmetal in a high-burnup cladding.

AXIAL SPLITTING IN HIGH-BURNUPCLADDING UNDER RIA CONDITIONS

Stress in the Hydride-Rich Metal Rim

The outermost region of metal beneath thethick waterside oxide of a high-burnup PWRcladding contains a high volume fraction ofmacroscopic hydrides aligned in thecircumferential direction. Even withoutmassive hydride blisters that form when oxide-

spalled fuel rods are operated continuously, thevolume fraction of the circumferentialhydrides can be as high as =80Y0. Then,considering the microstructural characteristicsdiscussed above, a complex state of residualstress will develop in such a hydride-richmetal-hydride composite. Although it isdifficult to deduce from commonly resolvedmicrostructural characteristics revealed byoptical microscopy, the effect of microscopic

hydrides, such as those in Fig.. 7, will beespecially important. The number density ofmicroscopic hydrides in the outermosthydride-metal composite in a high-bumupPWR cladding (fast fluence =9 x 1021 n cm-2)is likely to be a few orders of magnitude higherthan that shown in Fig. 7. That is, a large

number of microscopic hydrides will be locatedon the basal plane of the metallic regionbetween the macroscopic circumferentialhydrides in a high-burnup cladding. Thissituation is illustrated in Fig. 8. Because of thehigh density of microscopic hydrides on thebasal plane, basal slip will be inherentlydifficult to occur in the hydride-rich hydride-metal-composite layer. Consistent with this,no basal-slip dislocation channels wereobserved in any specimens that were preparedfrom the irradiated cladding (Table 1). Thisfinding is also consistent with the premise thatmicroscopic hydrides form during operation.

Under RIA-like loading situations of high-bumup cladding, e.g., in CABRI or NSRRreactor, the hydride-rich outermost metallicregion will be subject to several types ofstresses, which are illustrated in Fig. 8 and canbe summarized as follows:

1. Axial tensile stress due to the reaction forceassociated with axial constraint. Pellet-cladding-mechanical-interaction (PCMI)force is predominant in RIA situations forZircaloy high bumup cladding. Because ofthe tight binding between pellet and cladding,cladding contraction in the axial direction isprevented under PCMI loading. In theabsence of an axial constraint, however,axial contraction is the natural process whenthe textured cladding expands in thecircumferential direction. Therefore, areaction tensile force is produced in the axialdirection when a high-bumup fuel cladding

//

—. -. —— -. .. I

.

(A) Axial-Radial View .

.- ?7Stress at Crack-Tip :*.:~-

II

1!Metal near Hydride -‘ ----- -.

1. axial tensile stress due to constraint in axial elongatkk -2. axial shear stress due to hydride volume expansion3. axial tensile stress due to PCMI

4. axial tensile stress due to oxide volume expansion5. tangential shear stress due to hydride volume expansion

6. tangential tensile stress due to PCMI7. tangential tensile stress due to oxide volume expansion

8. tangential tensile stress due to fission gas

9. radial compressive stress due to hydride compaction

LJ

tl-.. -Ptol D.-%

Figure 8.

I

-a

of microscopic and macroscopic hydrides,Schematic illustration of distributioncrack propagation, and crack-tip stress components in metal near macroscopichydrides in hydride-rich outermost metallic region of high-burnup PWR cladding.

@-

2.

3.4.

5.

6.7.

8.9.

expands under PCMI force. This force canbe significantly large, comparable to thePCMI force itself.Axial shear stress due to hydride volumeexpansion. Because of volume expansion(Pilling-Bedworth ratio =1.16), a hydrideplate stretches significantly in the axial andtangential directions. As a result, a shearstress is produced in the metal in thesedirections. Shear stress is smaller in themiddle and larger near the edge of thehydride plate.Axial tensile stress due to PCMI.Axial tensile stress due to oxide volume “expansion.Tangential shear stress due to hydridevolume expansion.Tangential tensile stress due to PCMI.Tangential tensile stress due to oxidevolume expansion.Tangential tensile stress due to internal gas.Radial compressive stress due to hydridecompaction. When microscopic hydridesare highly dense, their tight compactionwill produce a significant level ofcompressive stress in the radial direction.

Particularly important is the influence ofshear stresses that are produced near a hydride.The shear stress will be more significant in theaxial direction whenever axial contraction o fhighly hydrided cladding is prevented orlimited, e.g., high-bumup fuel under RIA-likesituations, breached BWR fiel stressed byexpanding pellets, and spent-fuel claddingstressed by an expanding mandrel. Therefore,plastic deformation and crack propagation inthe axial direction under such circumstances isinherently facilitated, and as a result, longnarrow splits can be produced even at highspeed. In comparison, a free unconstrainedcladding tube under internal gas pressurizationwill develop relatively shorter and wider splits.

Evidence for crack propagation along themacroscopic hydrides, such as that depicted inthe tangential-radial plane in Fig. 8, is visiblein some optical photomicrographs of crosssections of high-bumup PWR cladding thatfailed under RIA-like conditions in the NSRR[17].

When microscopic hydrides are present inhigh density between macroscopic hydrides, asdepicted in Fig. 8, tearing of the narrow metalbetween the microscopic hydrides will occurrelatively easily, especially in the outermosthydride-rich region where residual stress andoxygen concentration are high. On amacroscopic scale, the fracture of the hydride-rich region will then appear like a propagationof a brittle crack, although fracture surfacemorphology could be either ductile,pseudocleavage, or mixture of the two types.

Hydride Redistribution . duringPreconditioning in Pulse Test

High-bumup Cladding REP-Nal (enthalpy atfailure =30 cal/g, pulse width =9.5 ms) andREP-Na8 (enthalpy at failure =83 cal/g, pulsewidth =40 ins), which fractured during pulsetesting in the CABRI reactor under RIA-likesituations [14,15,18], appear to exhibitmacroscopic hydrides with distinctivelydifferent morphology and distribution, whichwas imaged by optical microscopy. Thestructure of macroscopic hydrides in the as-irradiated “mother-rod” cladding of REP-Na 1was typical of a high-bumup Zircaloy-4 PWRcladding [14]; that is, predominantly

circumferential hydrides were concentratednear the waterside oxide/metal boundary. Thecladding midwall was nearly free of hydrides,and only small amount of circumferentialhydrides were located near the metal/pelletboundary. To show the radial distribution ofthe macroscopic hydrides of the mother rod,the REP-Na 1, and the REP-Na8 specimensmore quantitatively, the cladding wall in theoptical photomicrographs was divided into 10concentric rings. Then, the average distancebetween the circumferential hydrides wasmeasured as a function of radial distance fromthe waterside oxide/metal boundary. Therelative volume fraction of hydrides was thendetermined as a function of radial distance,assuming that all hydrides are platelets of thesame thickness. Figure 9 shows the radialdistribution of the hydrides in the threespecimens.

The hydride distribution in the cross-sectionalmetallographs of REP-Na8 is similar to the

,. .--— - . . ~-+r , ..,,,., - .—-- , -,———.——.

... . ..—. . .

#

Mother Rodcross Secflon Q02

I

g 20u

Before Test s

25~ 25

I

ri

g 15z>m; 10

zg=5=

~.a

0 0.5 1.5 2.5 3.5 4.5 5.S 6.5 7.5 8.5 9.!L-xE_J

Fractional Dlelance from Waterside OxidelMetal Boundary Fractional Distance from Waterside Oxide/hfetal Boundary .

Fig. 9.Radial distribution of circumferential macroscopic hydrides: (left) as-irradiated“mother-rod” sibling cladding of REP-Nal; distribution was measured along two radiallines, (right) REP-Nal and REP-Na8 after test; distribution was measured along sixand two radial lines, respectively.

40~ 1 I -1

❑ REP-Nal, cracks nucleated at OD and ID sides

W REP-Na8, cracks nucleated mostly at OD side

,,..,. ‘.

.,

,,, . .

,.

,. . . .

. .,,

,.-.

.,.. . ..

. .

Ot 1

OD metal ODhydride blister

hydride distribution in the mother rod (i.e.,typical hydride structure for PWR cladding,similar to that of HBO-series specimens testedin the NSRR [17]). Although the volumefraction of hydrides in the =1OYOof the wallnear the waterside oxide was close to 50°/0,hydride volume fraction in the =50Y0 of thewall near the pelletside was insignificant. Incontrast, hydride volume fraction in the =50°/0of the wall of REP-Nal near the pelletside was=3-5 times higher than that of REP-Na8.From these observations, we conclude thatsignificant dissolution and redistribution ofhydrides appear to have occurred during the

ID metal

,

Fig. 10.

Distribution of crackinitiation sites in REP-Na 1and -Na8 cladding.

4

test of REP-Nal, whereas redistribution ofhydrides in REP-Na8 was not significant.

The contrasting hydride structures of REP-Nal and -Na8 are likely to be associated withthe preconditioning procedure, i.e., the time-at-temperature in which the specimens werethermally equilibrated before pulse loading.According to available information [31], REP-Na8 was preconditioned at =310°C for =12 h,whereas REP-Nal was preconditioned at asignificantly higher temperature of=380°C for=13 h. At =380°C, significant amounts ofhydride in the as-irradiated material will bedissolved in the metal, and new hydrides will

,

I

precipitate as the metal is slowly cooled tothe pulse temperature of =290°C.

In the cross-section metallographs of REP-Nal and -Na8, a large number of incipient(partial) and a limited number of through-wallcracks were observed [14, 15, 18]. Cracknucleation sites can be classified into threetypes, OD metal, OD-side hydride blister, andID metal. In REP-Na 1, numerous incipientcracks were distributed more or less randomlyon the OD and the ID sides of the cladding,whereas in REP-Na8, incipient cracks wereobserved predominantly on the OD side. Thisdistribution is shown in Fig. 10. Thecontrasting crack distribution indicates thatmost of the cladding wall in REP-Nal wasbrittle, and as a result, cracks nucleated moreor less randomly at the OD and ID sides of thecladding. It appears that in REP-Na8, cracksnucleated mostly at the brittle OD metal orhydride blisters, whereas the ID side wasrelatively ductile, and hence, resistant tocracking.

Similar redistribution of macroscopic hydridesis absent in the HBO-series PWR claddingtubes tested in the NSRR [17]; this is expectedbecause the preconditioning and pulsetemperatures in those tests were =20°C.

Hydride Reorientation

Reoriented hydrides (i.e., macroscopic radialhydrides) are absent in HBO-series specimenstested at =20”C in the NSRR [17]. However,some degree of hydride reorientation isevident in the specimens tested in the CABRIreactor. As pointed out in Ref. 14, the degreeof reorientation appears to be morepronounced in REP-Nal than in -Na5. REP-Na8 also exhibited reoriented hydrides,although the number density was significantlylower than that in REP-Nal. Most reorientedhydrides in REP-Na5 appear to be in the=30Y0 of the wall near the OD oxide. Thetiming of hydride reorientation is, however,not clear, i.e., before or after the occurrenceof axial splitting. If it occurred before failure,axial splitting could have been significantlyinfluenced. Although it is not unambiguous,the fact that hydride reorientation wassignificantly more pronounced in REP-Na 1

than in -Na8 suggests that at least some radialhydrides in the former were produced beforefailure, i.e., when the tube was cooled from thepreconditioning temperature of =380°C to thepulse temperature of =290°C.

Radiation Anneal Hardening

Irradiated Zr and Zircaloys are susceptible tothe phenomenon of radiation-anneal

“hardening (RAH). The phenomenon refers tothe hardening of irradiated alloys when theyare annealed at a certain temperature range(=320-390”C) in the absence of neutron flux[32-34]. The phenomenon has been reportedfor both BWR and PWR cladding. It has beenattributed to O interaction with andsegregation to dislocations and defect clusters[19, 29, 35] or to irradiation-inducedprecipitation of O-rich phases [29]. It islikely that RAH occurs in high-bumup claddingunder pulse test conditions if thepreconditioning temperature reaches =320-390”C for a significant period of time, e.g., inREP-Nal. The effect is essentially adeleterious artifact effect, which shouldcontribute to a lower ductility of claddingspecimen and a lower enthalpy at failure.

Effect of Pulse Temperature Relative toDuctile-Brittle Transition Temperature

Ductile-brittle transition (DBT) of heavilyhydrided or irradiated Zircaloys [14,16,36] andZr-2.5Nb [37] has been investigated underconditions of impact [16], tensile [14], andcompact-tension fracture-toughness [36,37]testing. Under ring-stretch tensile testingconditions, Papin et al. reported a pronouncedeffect of high strain rate [14]. As discussed inRef. 16, the effect of pulse temperature as it isrelated to ductile-brittle transition temperatureis probably the most important factor thatsignificantly influences the result of asimulated RIA test on high-bumup fiel. Inthis respect, for specimens with similarmicrostructural characteristics, a pulse testconducted at =20”C, e.g., the tests in theNSRR [17], is probably a more conservativetest than a pulse test conducted at =290, e.g.,the tests in the CABRI reactor [14,18],because the lower-temperature test may placethe material below its DBT temperature.

/s

.— ——. ..—— -.———------

Effect of Pulse Width

The effects of pulse width” have beeninvestigated in detail by Meyer et al. [13].With regard to the relative failure behavior ofREP-Nal (pulse width =9.5 ms, enthalpy atfailure =30 cal/g) and REP-Na8 (pulse width=40 ms, failure enthalpy =83 cal/g), theeffects of the greater pulse width, i.e., highertemperature of cladding at the time ofmaximum stress and slower strain rate, areconsidered a major factor that contributed to ahigher enthalpy at failure for ~P1Na8.However, a direct conclusion about theseparate effect of pulse width is difllcult tomake, because, as shown in Fig. 9, the twospecimens exhibited contrastingmicrostructural characteristics that wereattributed to the significant difference inpreconditioning procedures.

1.

2.

CONCLUSIONS

Significant residual tensile stress is producedin the outermost metallic region inassociation with oxide volume expansion.The residual stress is an important factorthat promotes crack initiation in the outer-diameter side of irradiated fiel cladding.

Habit plane (i.e., {107}zr plane) and otherprecipitation behavior of macroscopiccircumferential hydrides in irradiatedZircaloy-2 and -4 fhel cladding areessentially the same as those of unirradiatedunstressed Zircaloys. At =300°C, hydrides inirradiated cladding are ductile, and brittlecrack of a hydride-itself does not occur.

3. In addition to macroscopic hydrides that arecommonly observed by optical microscopy,small “microscopic” hydrides are present inirradiated Zircaloy-2 and -4 cladding in highdensity; these can be resolved only bytransmission electron microscopy. Thehabit plane of microscopic hydrides is thebasal plane {002}zr, and the number densityis at least a few orders of magnitude greaterthan that of macroscopic hydrides in high-bumup cladding. Precipitationcharacteristics suggest that densemicroscopic hydrides play an important role

4.

5.

in axial splitting in high-bumup claddingunder RIA-like situations.

Complex residual stress is produced at theboundary between the metal matrix and ahydride in high-bumup cladding under RIAconditions. Large shear stress is produced inthe axial direction, especially when claddingis axially constrained because of pellet-cladding mechanical interaction. and tightbonding at high bumup. Because of the largeaxial shear stress, deformation and crackpropagation in the axial direction, i.e., axialsplitting, appears to be greatly facilitated inhigh-bumup cladding that containsmacroscopic and microscopic hydrides inhigh density.

Several factors associated with theprocedures for test preparation, specimenpreconditioning, and pulse initiation, i.e.,pulse temperature relative to ductile-brittle-transition temperature of the cladding, pulsewidth, hydride redistribution andreorientation during preconditioning, andradiation-anneal hardening, can significantlyinfluence the results of IUA testing of high-bumup fhel. Because of these factorssimulated RIA-like testing in a test reactorcould produce significantly conservative testresults for high-bumup fiel.

ACKNOWLEDGMENTS

The author is gratefid to R. O. Meyer, H. H.Scott, and M. C. Billone for helpful discussionsand materials, and L. J. Nowicki for preparingthe specimens for TEM analyses. This workwas supported by the U.S. Nuclear RegulatoryCommission, Office of Nuclear RegulatoryResearch.

1.

2.

/6

REFERENCES

Efsing, P. and Petterson, K., “TheInfluence of Temperature and YieldStrength on Delayed Hydride Cracking inHydrided Zircaloy-2,” in Zirconium in theNuclear Industry, 1Ith Intl. Symp., ASTMSTP 1295, 1996, p. 394.

Grigoriev, V. and Josefsson, B., “On theMechanism of Zircaloy Cladding AxialSplits,” J. Nucl. Mater. 257 (1998) 99.

I

I

,#

3,

4.

5.

6.

7.

8.

9.

10

Petterson, K.; Kese, K.; and Efsing, P.,“Studies on Delayed Hydride Cracking ofZircaloy Cladding,” in Proc. 9th Intl. Conf.on Environmental Degradation ofMaterials in Nuclear Power Systems -Water Reactors, August 1-5, 1999,Newport Beach, CA.

Edsinger, K.; Vaidyanathan, S.; andAdamson, R. B., “On the Mechanism ofAxial Splits in Failed BWR Fuel Rods,” inProc. 9th Intl. Conf. on EnvironmentalDegradation of Materials in Nuclear PowerSystems - Water Reactors, August 1-5,1999, Newport Beach, CA.

Lysell, G.; Grigoriev, V.; and Petterson, K.,“Characteristics of Axial Splits in FailedBWR Fuel Rods,” in Proc. 9th Intl. Conf.on Environmental Degradation ofMaterials in Nuclear Power Systems -Water Reactors, August 1-5, 1999,Newport Beach, CA.

Simpson, L. A.; and Puls, M. P., “TheEffects of Stress, Temperature, andHydrogen Content on Hydride-InducedCrack Growth in Zr-2.5%Nb,” Met. Trans.A (1979) 1093.

Puls, M. P.; Simpson, L. A.; and Dutton,R., “Hydride-Induced Crack Growth inZirconium Alloys.” Fracture Problems andSolutions in the Energy Industry, PergamonPress, New York, 1982, p. 13.

Coleman, C. E., “Effect of Texture onHydride Reorientation and DelayedHydrogen Cracking in Cold-Worked Zr-2.5Nb,” in Zirconium in the NuclearIndustry, 5th Intl. Symp., ASTM STP 754,1982, p. 393.

Chow, C. K.; and Simpson, L. A., “Analysisof the Unstable Fracture of a ReactorPressure Tube Using Fracture ToughnessMapping,” in Case Histories InvolvingFatigue and Fracture Mechanics, ASTMSTP 918, 1986, p. 78.

Cheadle, B. A.; Coleman, C. E.; andAmbler, J. F. R., “Prevention of DelayedHydride Cracking in Zirconium Alloys,” inZirconium in the Nuclear Industry, 7th Intl.Symp., ASTM STP 939, 1987, p. 224.

11. Coleman, C. E.; Cheadle, B. A.; CauseY, A.

12

13

14.

15.

16.

17.

18.

19.

R.; Chow, C. K.; Davies, P. H.; McM~nus,M. D.; Rodgers, D.; Sagat, S.; van Drunen,G., “Evacuation of Zircaloy-2 pressureTubes from NPD,” in Zirconium in theNuclear Industry, 8th Intl. Symp., ASTMSTP 1023>1989, p. 35.

Moan, G. D.; Coleman, C. E.; Price, E. G.;Rodgers, D. K.; and Sagat, S., “Leak-Before-Break in the Pressure Tubes ofCANDU Reactors,” Int. J. Pres. Vessel &Piping 43 (1990) 1.

Meyer, R. O.; McCardell, R. K.; Chung, H.M.; Diamond, D. J.; and Scott, H. H., “ARegulatory Assessment of Test Data forReactivity-Initiated Accidents; NuclearSafety 37 (1996) 271.

Papin, J.; Balourdet, M.; Lemoine, F.;Lamare, F.; Frizonnet, J. M.; and Schmitz,F., “French Studies on High-Burnup FuelTransient Behavior under RIA Conditions,”Nuclear Safety 37 (1996) 289.

Lemoine, F.; and Balourdet, M., “RIARelated Analytical Studies and SeparateEffect Tests, “ in Proc. 1997 Intl. TopicalMeeting on Light Water Reactor FuelPerformance, American Nuclear Society,La Grange Park, Illinois, 1997, p. 693.

Chung, H. M.; and Kassner, T. F., “CladdingMetallurgy and Fracture Behavior duringReactivity-Initiated Accidents at HighBurnup,” J. Nucl. Eng. and Design, 186(1998) 411.

Fuketa, T.; Nagase, F.; Ishijima, K.; andFujishiro, T., “NSRR/RIA Experiments withHigh-Bumup PWR Fuels,” Nuclear Safety37 (1996) 328.

Schmitz, F.; and Papin, J., “REP-Na 10:another RIA Test with a Spalled HighBumup Rod and with a Pulse Width of 30ins,” in: Proc. 26th Water Reactor SafetyInformation Meeting, NUREG/CP-O 166,vol. 3, U. S. Nuclear RegulatoryCommission, 1999, p. 243.

Chung, H. M.; Yaggee, F. L.; and Kassner,T. F., “Fracture Behavior andMicrostructural Characteristics of IrradiatedZircaloy Cladding.” in Zirconium in theNuclear Industry: 7th Intl. Symp., ASTMSTP 939, 1987, pp. 775-801.

n

. ,- ———. -=.. — _._.. _.—

20.

21.

22.

23.

Garde, A. M., “Effects of Irradiation andHydriding on the Mechanical Properties ofZircaloy-4 at High Fluence,” in Zirconiumin the Nuclear Industry, 8th Intl. Symp.,ASTM STP 1023, 1989, p. 548.

Smith Jr., G. P.; Pirek, R. C.; Freeburn, H.R.; and Schrire, D., “The Evaluation andDemonstration of Methods for ImprovedNuclear Fuel Utilization,” DOE/ET/340 13-15, CEND-432, ABB CombustionEngineering, 1994, pp. 4-60 to 4-73.

Garde, A. M.; Smith, G. P.; and Pirek, R.c., “Effects of Hydride PrecipitateLocalization and Neutron Fluence on theDuctility of Irradiated Zircaloy-4,” inZirconium in the Nuclear Industry, 11thIntl. Symp., ASTM STP 1295, 1996, p.407.

Newman, L. W., “The Hot CellExamination of Oconee 1 Fuel Rods AfterFive Cycles. of Irradiation,”DOE/ET/342 12-50 UC-78 BAW-1 874,Babcock & Wilcox, 1986.

24. Donaldson, A. T., “Growth in Zircaloy-4Fuel Clad Arising from Oxidation atTemperatures in the Range 623 to 723 K,”in Zirconium in the Nuclear Industry, 9thIntl. Symp., ASTM STP 1132, 1991, p.177.

25. McMhm, A.; Darby, E. C.; and Schofield, J.S., “The Terminal Volubility of Hydrogenin Zirconium Alloy s,” in Zirconium in theNuclear Industry: 12th Intl. Symp., June15-18, 1998, Toronto.

26. Langeron, J. P.; and Lehr, P., Compt.Rend. 243 (1956) 151.

27. Langeron, J. P.; and Lehr, P., Rev. Met. 60(1958) 901.

28. Westlake, D. G.; and Fischer, E. S., Trans.TMS-AIME 224 (1962) 254.

29. Westlake, D. G., “The Habit Planes ofZirconium Hydride in Zirconium andZircaloy,” J. Nucl. Mater. 26 (1968) 208.

30. Chung, H. M., “Phase Transformations inNeutron-Irradiated Zircaloys,” inRadiation-Induced Changes inMicrostructure, 13th Intl. Symp., ASTMSTP 955, 1987, p. 676.

31.

32,

33.

34.

35

Balourdet, M., private communications,.

1999.+

Snowden, K. U.; and Veevers, K.,“Radiation Hardening in Zircaloy-2,”Radiation Effects 20 (1973) 169. 1

Onchi, T.; Kayano, H.; and Higashiguchi,Y., “The Inhomogeneous DeformationBehavior of Neutron Irradiated Zircaloy-2,” J. Nucl. Mater. 88 (1980) 226...Onchi, T.; Kayano, H.; and Higashiguchi,Y., “The Inhomogeneous PlasticDeformation and Its Relevance to IodineStress Corrosion Cracking Susceptibility inIrradiated Zircaloy-2 Tubing,” J. Nucl.Mater. 116 (1983) 211.

Adamson, R. B.; and Bell, W. L., “Effectsof Neutron Irradiation and Oxygen Contenton the Microstmcture and MechanicalProperties of Zircaloy,” in Microstructureand Mechanical Properties of MaterialsVol. 1, p. 237, Proc. Intl. Symp., X’ian,Republic of China, 21-24 October 1985.

36. Davies, P. H.; and Steams, C. P., “FractureToughness Testing of Zircaloy-2 PressureTube Material with Radial Hydrides UsingDirect-Current Potential Drop,” in FractureMechanics Vol. 17, ASTM STP 905, 1986,p. 379.

37. Wallace, A. C.; Shek, G. K.; and Lepik, O.E., “Effects of Hydride Morphology on Zr-2.5Nb Fracture Toughness,” in Zirconiumin the Nuclear Industry, 8th Intl. Symp.,ASTM STP 1023, 1989, p. 66.