Characteristics and Use of Urania-Gadolinia Fuels

IAEA-TECDOC-844

Characteristics and use of urania-gadolinia fuels

INTERNATIONAL ATOMIC ENERGY AGENCY The IAEA does not normally maintain stocks of reports in this series. However, microfiche copies of these reports can be obtained from

INIS Clearinghouse International Atomic Energy Agency Wagramerstrasse 5 0 10 P.Ox Bo . A-1400 Vienna, Austria

Orders should be accompanied by prepayment of Austrian Schillings 100, in the form of a cheque or in the form of IAEA microfiche service coupons which may be ordered separately from the INIS Clearinghouse. originatine Th g Sectio f thino s publicatio IAEe th An i was: Nuclear Material Fued san l Cycle Technology Section International Atomic Energy Agency Wagramerstrasse5 0 10 x P.OBo . A-1400 Vienna, Austria

CHARACTERISTICS AND USE OF URANIA-GADOLINIA FUELS IAEA, VIENNA, 1995 IAEA-TECDOC-844 ISSN 1011-4289 © IAEA, 1995 Printe IAEe th AustriAn y i d b a November 1995 PLEASE BE AWARE THAT MISSINE TH AL F LO G PAGE THIN SI S DOCUMENT WERE ORIGINALLY BLANK FOREWORD

In BWRs, the issue of power mismatch between fresh fuel assemblies and partially burned fuel assemblies causing the power in the former to be depressed was recognized at an early stage. Since the earlier absorbers used to address this issue presented inconveniences, the use of gadolinium in urania-gadolinia fuel became routine in all BWRs soon afte decisioe th r f Generano l Electri implemeno ct Dresden mid-1960se i t th i t n i n2 . fueR l vendorBW e th l sAl have since progressively adopte dtotae gadoliniath f f o l o t Ou . 420 reactors around the world with an installed capacity of 330 GW(e), approximately 90, wit installen ha d capacit GW(e)0 7 f BWRsyo e ar , . Experience with gadolinia ove pase rth t twenty years has become statistically significant, providing confidence in the concept.

When advanced duty conditions (i.e. extended burnup, increased reactor cycle lengths and/or in-out core refuelling schemes) were considered and implemented in PWRs, f burnablo e us e th absorber r thasfo t reactor type also came into consideration. Theres i an obvious advantag t sacrificinno n ei fuee positiond gth ro l r burnablsfo e absorberso s , integral burnable absorbe attractivn a r fues wa l e choice gooe .Th d experience accumulated with gadolini face ath tfued thaan l t some fuel vendors were fabricatind an g R bothBW PWR fuel made gadolini prime ath e choic meeo et needse th t .

recene Inth t past extendee ,th d dutie fuelR ,sPW considerer fo d an R d botBW r hfo an perceivee dth d advantage usinn si g burnable absorber WWERn s i same th r esfo reasons aPWRn i s s increase e prospectdth r gadolinisfo a utilizatio hastened nan e needth r dfo higher gadolinia contents. Against this background IAEe ,th A considere t dappropriati o et launch in 1990 a Co-ordinated Research Programme on Technology and Performance of Integrated Burnable Absorbers for Water Reactor Fuel (in short "BAF"). The aim was to issue a report summarizing the various aspects of burnable absorber utilization. The participation was on a voluntary basis. This report is the result of the inputs of the participant Co-ordinatee th o st d Research Programme.

The IAEA would like to express its appreciation for the work performed by the participants and, in particular, to acknowledge the contributions of H. Bairiot (Belgium), D. Farrant (United Kingdom . OnufrieV d an ) v (Russian Federation e draftinth n d i ) gan review of this report.

The scientific secretaries of the programme were successively P. Chantoin and G. Sukhanov of the IAEA, Division of Nuclear Fuel Cycle and Waste Management. EDITORIAL NOTE

In preparing this publication for press, staff of the IAEA have made up the pages from the original manuscript (s). The views expressed do not necessarily reflect those of the governments of the nominating Member States or of the nominating organizations. Throughout the text names of Member States are retained as they were when the text was compiled. The use of particular designations of countries or territories does not imply any judgement by publisher,the legalthe IAEA, to status the as of such countries territories,or of their authoritiesand institutions delimitationthe of or theirof boundaries. The mention of names of specific companies or products (whether or not indicated as registered) does implyintentionnot any infringeto proprietary rights, should construednor be it an as endorsement or recommendation on the part of the IAEA. CONTENTS

INTRODUCTION ...... 7 .

1. STRATEGIC CONSIDERATIONS FOR THE USE OF BURNABLE ABSORBERS .... 9 1.1. Introduction ...... 9 1.2. Applications of burnable absorbers ...... 9 1.3. Advantages and disadvantages of burnable absorber fuel ...... 10 1.4. Choice f integrateso d burnable absorbers ...... 2 1 .

. UNIRRADIATE2 D GADOLINIA FUEL PROPERTIES ...... 7 1 .

2.1. Structure and chemistry of the (U,Gd)O2 solid solutions ...... 17 2.2. Physical properties ...... 3 2 . 2.3. Mechanical properties ...... 0 4 . 2.4. Neutronic properties ...... 44

3. FUEL MANUFACTURING ...... 0 5 . 3.1. Fabrication processes ...... 50 3.2. Quality control ...... 53

4. DESIGN AND MODELLING CONSIDERATIONS ...... 59 4.1. Principle f gadoliniso a fuel rod, assembl reactod yan r core design ...... 9 5 . 4.2. Modelling ...... 5 7 .

. EXPERIENC5 E WITH GADOLINIA FUEL ...... 3 8 . 5.1. Utilizatio f gadolinino a fuel ...... 3 8 . 5.2. Nuclear dat Neutronia— c measurements ...... 0 9 . 5.3. Thermal-mechanical behaviour ...... 2 9 . 5.4. Behaviour under reactivity initiated accidents ...... 99

. FUE6 L CYCLE BAC...... D KEN 3 10 .

7. CONCLUSIONS ...... 105

APPENDI : XI GADOLINIU ERBIUD REVIEA MAN M- RESOURCE F WO S AND SUPPLY ...... 107 APPENDIX II: GADOLINA FUEL UTILIZATION IN FRANCE ...... 114 APPENDIX III: GADOLINIA FUEL UTILIZATION IN JAPAN ...... 121 APPENDI : GADOLINIXIV A FUEL UTILIZATIO REPUBLIE 1 TH NN 15 KOREI F CO . .. A APPENDI : XGADOLINIV A FUEL UTILIZATIO BELGIUNN I M ...... 7 17 .

ABBREVIATIONS ...... 5 18 .

PARTICIPANTS IN THE CO-ORDINATED RESEARCH PROGRAMME ...... 191 INTRODUCTION

Burnable absorber fuels (BAF utilizede beine ar ) ar r g,o considere r utilizationdfo n ,i l BWRsal mosn i , t PWR mord san e recentl WWERsn yi topie thereforTh s c.i e relevano t approximately 330 out of the 420 operating reactors in the world, representing 280 of the GW(e0 33 ) installed capacity worldwide lighe f th thi o t n sI . importance IAEe s th , Aha decide issudo t e this report providin overaln ga lvariou e vieth f wo s aspect f BAFso . With e exceptioth f Chapte no e whol th , e1 r repor devotes i t urania-gadolinio dt a fuel ("Gd fuel")1, the most commonly used BAF, and a comprehensive technical review of this topic provideds i , althoug repore hth t doe t includsno completea e surve l examplesal f yo d G f o utilization throughou industrye th t repore Th organize.s i t followine th dn i g way:

Chapter 1. Strategic considerations: describes the reasons for application of BAFs, their advantage disadvantaged san varioue th d ssan absorbers being utilized.

underlyine Th larga o t eg e extensucces fuedu d s G almoso i lt t f so 0 3 t years industrial utilizatio BWRsnn i applicatioe Th . PWRno t mors i e recent, as more challengin coresgR condition addition I PW . n i t no t me s e havb o et Gd fuel alternativo ,tw typeF eBA s have been develope utilizee ar d dn i dan commercial PWRs. They present advantages and disadvantages which are compare thin di s chapter.

Chapter 2. Unirradiated gadolinia fuel properties: outlines the differences between Gd fuel propertie fueU d l propertiessan .

homogeneousls i Eved G nf i y disperse uraniue th dn i m oxide matrix t woul,i d normally be expected to affect the properties of the fuel for three reasons. Firstly naturae th , l structur oxid d body-centerea G s ef i e o d cubic lattice, while U oxide is a face-centered cubic lattice. Secondly, Gd is a trivalent atom, unlik whic, eU tetravalenhs i nuclean i t r fuel. Thirdly absorptioe ,th n cross sections of the various Gd isotopes interfere with the resonances of the U isotopes. Both of the first two differences influence or are likely to influence the physical and mechanical properties of BAF in "as-fabricated" conditions. Moreover, the fabrication technique influences the dispersion of Gd within the fuel.

Chapter 3. Fuel manufacturing: deals with the specific fabrication processes and quality control techniques for Gd fuel.

Although fabrication techniques to produce perfectly homogenous Gd fuels have been developed l industriaal , l fabrication processe basee a sar n do mechanical blending of the two ingredients. Specific quality controls are implemented to verify the proper dispersion of Gd within the BAF fuel and specifiee th d fuepositioninF lfue e BA rod th e l n sth assemblyi f go .

Chapte . 4 r Desig modellind nan g considerations: describe specifie th w csho properties of Gd fuel and the manufacturing constraints are taken into account for utilization of Gd in BWRs, PWRs and WWERs and how the modelling codes incorporate Gd fuel properties.

1 urania-gadolinia fuel is usually called "gadolinia fuel" within the nuclear industry. In this report, for convenience, the abbreviation "Gd fuel" is being used. Design flexibility is afforded by the use of Gd: the location of the Gd fuel rods within the fuel assembly and the axial shaping of the Gd along the Gd fue tunele rodsb varioun do t ,ca s design objectives (e.g. reduced fabrication cost, improved design margins, improved reactivity evolution and so on). fuemorAa d sG s i l e complicated fuel than standar fueld U desig e ,th n code incorporates a number of modifications to address the use of Gd fuel and licensings it .

Chapter 5. Experience with qadolinia fuel: presents typical examples of Gd fuel utilizatio resultd nan s from corfued ean l surveillanc frod mean experimental programmes.

Extensive experience with Gd fuel acquired by several fuel vendors and fued G l meetino t d utilitiele gs higsha h performance standardsw fe A . examples illustrate extent of this experience.

Chapter 6. Fuel cycle back end: considers the implications of the presence of Gd in spent fuel.

An obvious consequence of additions of Gd to the fuel is to increase the quantity of material to be incorporated into high level waste in the reprocessing option. 1. STRATEGIC CONSIDERATIONS FOR THE USE OF BURNABLE ABSORBERS

1.1. INTRODUCTION

The need to improve reactor performance through longer cycle lengths or improved fuel utilizatio bees nha n apparent sinc e beginnineth f commerciago l nuclear power generation. Among several modifications introduced as a consequence, the fuel initial enrichmen bees ha tn increased, whic means hha t thaadditionae th t l amoun f fissilo t e material in the core has had to be compensated for by the introduction of additional absorber material in the core. This was initially achieved by lumped absorbers located between the assemblies (control rods or removable steel curtains) or by soluble absorber (boric acid) in the coolant.

f solublo e BWRsn I us e e absorbeth , coolante th n i r / moderato prohibites wa r r dfo technological reasons. Therefore, lumped absorbers were the only solution. Since the earlier lumped absorbers (steel curtains) were causing not only the power in the fresh fuel assemblies depressede (FAb o t ) t als e poweadjacene bu ,o th th n i r t assembliese th , advantage of locating the absorber within the FA became obvious. Due to the high rating fuee th lf rodo s (FR), withdrawin replaco t R F gt ea witi unfuellen ha d absorbes wa d ro r not possible. This is the reason why integral burnable absorbers were introduced in 1967 Dresden i . nThi2 s particular burnable absorbe gadolinius wa r m oxide mixefuele th .n di Since then, this type of absorber has been routinely used in this type of reactor. It is routinely called "gadolinia fuel wilabbreviated e b lan " fueld thi"G n i " so dt report.

f solublo e PWRsn I us e e absorbeth , coolane th acceptables n i r wa t . Boric acidn i the coolant/moderator has therefore been used routinely to meet the needs of increased reactivity compensation. However, the increase in initial fuel enrichment cannot be indefinitely compensate increasiny b r dfo g boric acid concentration beyons a , certaida n concentration, thermal expansio f wateno t startura p reduce quantite sth f boroyn o i ) n(B the core, resulting ultimately in a positive moderator temperature reactivity coefficient (MTC). To avoid this unacceptable situation the introduction of solid burnable absorbers consideredfues e th wa l n i alternativo Tw . e solutions have been successively developed:

Introduction into the FAs of unfuelled rods containing burnable absorber (borosilicate glass, boron carbide disperse aluminn di othersd - aan so n i ) calle Clusted dRo r Control Assemblies whic withdrawe har nA F fro e mth after its first irradiation cycle.

Addition of an absorber within the fuel rods themselves.

1.2. APPLICATION BURNABLF SO E ABSORBERS

1.2.1. Cycle length extension

To minimize the fuel cost, the amount of spent fuel and the amount of waste after reprocessing, increased fuel discharge burnup is considered as one of the most important goals by utilities. This has been confirmed by the results of the IAEA's programme WREBUSn.1].

differeno Tw t approache usee b increasdo n t sca e bur dependinp nu relative th n go e proportion of nuclear generation in the total installed electrical capacity of the country. The first is typically the type of management used in France where more than 70% of the electricity produce f nucleao s di r origin highee Th . r burnu achieves pi increasen a y db d fractioning of the reload, keeping reactor cycle length unchanged (annual refuelling scheme). The power mismatch between FAs of successive reloads can be kept under control withou burnabld neee ad th t o dt e absorbe frese th ho t rfuel .

For most other countries, wher proportioe eth f installeno d nuclear capacits i y low, increasing burnu increasind pan g cycle lengt joine har t goals orden I . achievo t r e these, burnable absorbers usee havb do t esinc e poweeth r mismatch betweee nth assemblies of successive reloads becomes too great to be compensated through in-core fuel management only.

1.2.2. Low leakage fuel management and vessel fluence reduction

The traditional PWR and WWER fuel management scheme (without burnable absorber loao t s di ) fresh fuel assemblie core th e t sa peripher movd yan e them toware dth centre of the core after the 1st cycle (out-in management).

To minimize irradiation damage to the vessel, an alternative fuel management scheme is now usually adopted consisting of fresh FAs loaded in the centre of the core whic thee har n shifte core th edo t peripher their yfo r last cycle (in-out management). This type of fuel management scheme is called a "low leakage loading pattern" (LLLP) and requires that the power of the fresh FAs be depressed: i.e. burnable absorbers must be introduce thosn di e FAs.

It is worth noting that, with the utilization of MOX fuel, higher fast neutron doses are produced compared wit fuelhU thereford an , e loadinn outei s r FA g fresX hMO avoidede positionb o t s , sha givin additionan ga l incentiv in-oue th r te fo reloa d strategy.

1.3. ADVANTAGE DISADVANTAGED SAN BURNABLF SO E ABSORBER FUEL

1.3.1. Power distribution control

The consequence of introducing new fuel assemblies into a core which has reached equilibrium is to induce mismatches in the power distribution.

In a reload without burnable absorbers, the core configuration is determined so as to be within the maximum permissible power peaking factor at the beginning of cycle. This peaking factor decreases with burnup.

For a reload incorporating BAF, the maximum power peaking no longer occurs at the beginning of life in a fresh subassembly, but rather after consumption of the burnable absorber. This effect is shown in Fig. 1.1 which presents the situation for Gd fuelled WWERs figure Th . e show coefficiene sth f poweo t r density non-uniformity withiA F e nth

(K, called assembly-wise peaking facto PWRs)n i r same th , e coefficien e corth er fo t q volume (Kv, called total peaking factor for PWRs) and the evaluation of the axial offset (AO). This figure shows that meetin desige gth n beginnine limitth t sa f lifgo e (max

K = 1.35 and max K = 1.56) does not imply that the criteria are met over the whole cycle. v

Inq fact situatioe th , n deteriorates afte EFPD0 20 r . Bot desigA h F cor d nean management are more complicated if burnable absorbers are utilized.

1.3.2. Waste volume reduction

As mentione doptioF earlier BA f lumpeno e replacee th , us d e absorbersth s , consisting either, for BWRs, of stainless steel curtains withdrawn after one cycle or, for PWRs, burnable contro assemblied ro l extracted an ss inserte FA fresde C th afte hdn RC i r their first irradiation cycle.

10 1.35-

1.30-

0 20 0 10 300 Time (EFPD)

Fig. 1.1. Typical time dependence variation the power of the of density non-uniformity coefficients (K^ and Kv) and of the axial offset (A^) in a WWER designed [1.2].Gd of use forthe

Today, with increasing emphasis being place reducinn do g wastes reductioe th , n of structural core materia discontinuiny b l stainlese gth s steel curtain optio BWRsnn i d ,an abandonmene th burnable th f o t e contro assemblied ro l PWRa pointsn i e ,ar favousn i f o r f BAFo e . us e th

For PWRs, these argument t vali countrier no dfo e sar s having take irreversibln na e decisio finae th ln ndisposao wastea s f a spenthi n o lI s .s FA tcase spene th , t burnable control rod assemblies can be inserted in the spent FAs, which act as a final repository for the thereb d t increasman no o ytotad e eth l volum f wasteeo .

1.3.3. Uranium / Wate r ratio

Lumped burnable absorbe introducee b n ca reactoa r n di r eithe y occupyinb r g control rod positions within a FA (PWRs and WWERs) or the water gap between FAs (BWRs replaciny b botr n I o )withi R . hF casesga FA n a moderatio e th , n rati modifieos i d from wha originalls wa t y designeoptimumn a e b o dt .

In the first case, the positioning of the lumped burnable absorbers is restricted to core position t locatesno contron do drivd ro l e mechanism locations.

In the latter case, the heat transfer surface between fuel and coolant is reduced due replacemene th absorbinn o t a y activn b a d f o gtero rod. This lead higheo st r heat flux hot-spot-factors.

11 These disadvantages of lumped burnable absorbers are not apparent, or are minimized f burnabli , e absorber incorporatee sar d within FRs.

1.3.4. Residual absorption penalty

Neutron captur absorben ei r isotopes isotopesresultw ne n si macroscopie th , c cross-section f whico s h shoul smals a possibls e da b e l absorbe e th lefe th b n i t f ei o t s i r FA up to end of life.

1.4. CHOICE INTEGRATEF SO D BURNABLE ABSORBERS

1.4.1. Gadolinia (Gd2O3)

progressive Th e developmen burnabla s a d G e f absorbeo t ever o ' n Gd r o frow/ m1 a fraction of a percent in the early utilizations in BWRs, to the present 6 to 10 w/o Gd level for PWRs enables ha , gooda d data buil e bas. Thib tup o e t s large dat gooe a th bas dd ean performanc wers eFR reasond G progressivs f it e o r sfo e utilizatio mosy nb t fuel suppliers.

However, with increasin contentd gG complexite ,th f nucleayo r design increases significantly and the fuel properties (in particular thermal conductivity and melting point) are affected to a significant extent. As a result, to maintain adequate safety margins, the muss desig reduce e poweFR e b t d th nG d powerd e an sharin th f o r d g an betwee s FR d nG standard uranium FRs becomes altered. In extreme cases, depleted U was utilized in early PWR application f hig o scontend o preven t h G F e erosiouncertainty BA th tt an f o n y margins in the design and licensing phases. With the development of data bases up to higher Gd contents, such conservative approaches are not necessary and the potential benefits of Gd fuel can be used to their full extent.

Burnable absorbers work by neutron capture on a large cross section isotope (eg.Gd155 and Gd157). The isotopes produced by this nuclear reaction (eg. Gd156 and Gd158) hav smalea l absorption cross section, whic undesirablhs i e since this induce reactivitsa y penalty ove life efuelth re timth f .eo

These difficultie implementinn si g hig contend hG t fuel have resulte otheo tw rn di alternative absorbers being applied commercially: ZrB2 and Er.

1.4.2. Zirconium diboride (ZrB) 2

Developed by Westinghouse under the name of IFBA (Integral Fuel Burnable

Absorber), this fuel consists of a thin layer of ZrB2 (0.02 mm) deposited by sputtering on 10 10 the surface of the UO2 pellets. The resulting B loading is 1.7 mg B /cm and the layer adheres perfectly to the U0 substrate. The material properties (including thermal conductivity) of the substrate 2 are not affected by this layer.

IFBA lowea present d advantagee ran th residua l F sal BA a f l so absorptio n penalty

0 than the other BAFs, since it can be designed for complete depletion of the Bat1 the end of the first irradiation cycle. Technical details on IFBA are given in [1.3 - 1.5]

Although apparently more complicated e fabricatioth , n techniques have been incorporated into a large scale manufacturing facility by Westinghouse. Issues such as the

* In this report, all the Gd contents are expressed in weight percent Gd0 in the fuel, but 3

noted w/o Gd, for simplification (see all abbreviations and symbols) 2

12 hygroscopic nature of the ZrB2 coating are simply addressed and the manufacturing and quality control complexit t substantiallno e ar y y different fro md fue G thosl r fo e manufacture.

Additionally, although there is a higher tritium content of the IFBA fuel (resulting fro e interactiomth f fasno t neutrons with )Bnegligibla , 10 s thiha s e impace th f o t reprocessing option. There is almost no environmental impact or Intermediate Level Waste impact and the High Level Waste impact is roughly the same as that from an equivalent mas f gadoliniaso .

The consumptio generateB f no whice dH h result neea n sreduco i dt e preeth - pressurization level of the fuel rod (FR). It implies an increased creep down rate of the cladding earl lifn yi e and, consequently, potentially affect propensityI sPC . Althouge hth production of He during B depletion presents no operational limitations, the He release mus explicitle b t y treate fuen ddesigni d ro l . This analysi readils si y performed witha properly benchmarked fuedesigd ro l n code.

desigA F e bases n i radia Th a axiad n do an l l shapin absorbere th f go radiae Th . l positioning of the IFBA FRs in the FA, typically 30-40 % of the FA, is designed to minimize

the radial power peaking within the FAs. The axial zoning is designed to reduce the axial

0 shape factor by avoiding the presence of Bat1 the FR ends, thus suppressing the power in the centre of the FR. The borated length in the pellet stack is typically 2.3 m over an active fuel length of 3.7 m.

Sinc absorbelayee e eth th thi s d i r nan r quite weak depressioe ,th thermae th f no l neutron flux within the absorber is low and the nuclear design is less complicated than for other BAFs. Unlike other BAF FRs, the IFBA FRs need not systematically be derated to account for lower thermal conductivity, lower melting point and increased nuclear design uncertainty margins.

Following irradiatio f tesno t segment3 BR Turken s i n i f s teso y FR d tPoin an 3 t since 1981, IFBA fuel has now been commercially utilized by Westinghouse since 1987. In 19904 reloatota3 a f , o dl batches incorporatin 0 werIFBs 00 FR A e 5 bein16 g g irradiate PWRs0 2 dn i , mos f theo t m operatin mont8 1 n gho reactor cycles.

Further developments bein f enricheo ge considere0 insteaus 1 dB e th f do e dar natural B and an increase in the B10 content of the IFBA FRs, resulting in a reduction of the proportion of IFBA FRs in a 1 7 X 17 FA from 18-48 % to 1 2-30 %. The within-FA power peaking factors could thu withou A reducee F s b leve e a th f o tl o dt IFB A e.g.:

peaking factors for FA with no IFBA: 1.05 - 1.04 peaking factorwitA F hr currensfo t design IFBA: 1.10- 1.04 peaking factors for FA with enriched boron IFBA: 1.05 - 1.04

It should be noted that local peaking factors can be lower with IFBA than for an assembly without IFBA dependin e specifith n o gc burnable absorber placementU , enrichmen latticd an t e etc.

A comparison of IFBA and Gd BAF by Westinghouse for a particular scenario indicates that:

cycle th e lengt increases hi witlowe% e th h1 o IFBAy rt db residuae sdu l absorption penalty

radiae th l power peaking facto whe % reduces i r 2 n 2. usin y db g IFBAs

13 the BOC HFP critical B concentration is increased by 80 ppm (i.e. 6%) when using IFBAs, leading potentially to a positive MTC. This is the only disadvantage state y Westinghouseb d , althoug a hnumbe f planto r e ar s operating well with appropriate MTC control using only IFBAs.

1.4.3. Erbium (Er)

effective Th e capture cross welsectione ar lr doublknowe E f th s o d enan resonance peak at 0.5 eV plays an important role in managing the MTC. This is the major advantage over E f r o ZrB. Otherwise advantagee moss th f ha o tt i latter,e th f so , succrosa s ha s

sectio2 n similar to B. The difficulties typical of Gd, i.e. neutron spectrum hardening and wide neutron flux variations wit depletiond hG therebe ar , y avoided [1.6].

s beeEha rn utilize n TRIGi d A research reactors since s bee197ha n d 4an

demonstrated to be predictable and reliable. For PWRs, Er0 is directly blended with UO 3 2

BAFsd G t onl a r proportioa require bu w/os ,s2 fo n yi numbe e o FR t r .Th 1 E d f f no o r 2 is typically 20 - 30 % of the FA although more recent verbal presentations quote higher percentages.

advantagee Th arer E f ,so besides those IFBA d commoan d : G o nt

a smaller impact on the fuel properties (thermal conductivity, melting point) due to the lower rare earth contents compared with Gd fuel and the absence of a thermal barrier at the FR surface (unlike the IFBA).

poweA F e r th excellen sharingd an C .t MT contro e th f o l

e maiTh n disadvantag higa s hi e residual poisonin e sloth wg o t effec e du t consumptio. Er f no

The adoption of Er by CE (now ABB/CE) was preceded by an experimental programme demonstrating tha propertiee th t fuepropertier e close E th ar f l so eo d t G f so benefin ca fue d databased t an G lfro e mth . Four demonstratio fabricates nFA 198n di 9 have been loade Calvern di tfou d Cliffan r sother2 Onofrn 199n si confirSa o t n 1e2 i m the validity of the design package and the fuel performance. The first full reload with Er is scheduled to be loaded in 1994. More recent brochures from ABB suggest that Gd fuel mors i e appropriat routine th r efo e cycle length fue r PWRsn i E l d morsan e appropriatr efo the very long cycle lengths, e.g. 24 months.

It should also be mentioned that Dysprosium (Dy) was initially contemplated as a burnable absorber in Belgium [1.7], before Gd was chosen due to its general commercial implementation which provided greater confidence. These early R & D and irradiations in Belgium have proved that Dy behaves, from fabrication and fuel properties viewpoints, similarly to Gd (and consequently Er).

1.4.4. Compariso f BAFno s

Based on an evaluation of relevant publications, a comparison of the three types of BA summarizeFs i Tabln di takin1 e1. g into accoun mose th t t common specificationr sfo PWR fuel.

The availability of raw materials to meet the increased demand that would result from the wide - spread utilization of any of these three solutions in nuclear power plants criterioa t (NPPno f theselecs o no i t ) e m on tpreferentially t onl No thers yi . ampln ea e

14 TABLE 1.1. COMPARISON OFBAF TYPES

BAFType Gd IFBA Er

Fuel fabricators (a) Wh(USA) AB (USAE B/C )

BAF concentrations o w/ 0 1 - 6 — 1-2.o [1.8w/ 5 ] fuen i l pellet

Thermal conductivity significant negligible small decrease

Melting point reduction significant negligible small

Proportion of BAF 3-6% 30 - 40 % % 0 3 - 0 2 FRs per FA

235 Reduction of U small negligible small quantity in the FA

Depletion rate high high medium

Residual reactivity low negligible significant

FA power distribution good fair (b) good

Local power peaking significant (c) low low

MTC control good fair best

parallel small additional large parallel Fabrication facility throughput throughput line throughput line equipment

Fuel reprocessability good questionable no problems anticipated

(a) FBFC (Belgium), SIEMENS (Germany), Fabricazione Nucleari (Italy), JNF, MNF and NFI (Japan), KNFC (Korea), ENUSA (Spain), ABB Atom (Sweden) BNFL (United Kingdom), TENEX/Novosibirsk/Ulbinskij (Russia/Kazakhstan), (USA)C SP BWFCd . an E G ,

faso t te depletiodu (b) n

desigo t e ndu challenge(c) s resulting from self-shieldin neutrod gan n spectrum effects

15 supply of B, but Er and Gd resources are also plentiful (App I). Taking Gd as an example, upper bound requirementd G f so calculatee b n sca : das r BWRfo s e Gdg GW k . 20 a 3/ 16 " " 0 5 forWWERd an s 33 " " for PWRs (assumed to operate on an 18 month cycle)

The total requirement worle refueo st th l dal l LWRs wit fued hG + l woul2 1 e db

1 + 6 = 19 t Gd2O3 / a, which represents only 1 to 3% of the annual production of Gd (App.I). If Er were to be the absorber used in refuelling all LWRs world - wide, however, this would represent an increase of 20 to 50 % of the present annual production of Er.

REFERENCE CHAPTEO ST R1

[1.1] INTERNATIONAL ATOMIC ENERGY AGENCY, Technical Report Series No. 343, Water Reactor Fuel Extended Burnup Study (WREBUS), 1992

[1.2] ONUFRIEV, V.D., PROSELKOV, V.N., "Issue f safetso y ensuranc economid ean c indice USSe th f Rso water-cooled reactor using U-Gd fuel", presente IAEAe th t da BAF Research Coordination Meeting, 22-25 October 1990, Vienna, Austria

[1.3] SIMMONS, R.L t al.e . , Integral Fuel Burnable Absorbers with 2rB PWRsn i ,

Nuclear Technology, Vol. 80, March 1988 2

[1.4] PRITCHETT, J.E t al.e . . Operational Experience with ZrB Integral Fuel Burnable

Absorbers, ANS 1987 Winter Meeting Proc., pp. 112 7 - 118

[1.5] SRINILTA, S. et al., A comparison of Gd and ZrB2 as Integral Fuel Burnable Absorber PWRsn si 198S 7AN , Winter Meetin - 126 g4 Proc.12 . . pp ,

[1.6] JONSSON , InitiaA. , l Physic PWRa n i s ,A EvaluatioB s a r E f no ANS Annual Meeting, Nashville, June 1990

[1.7] MESTDACH, G., FLIPOT, A.J., Control autobiographique de la répartition de

Dy203 dispersé dans une matrice en U02, Energie Nucléaire, mar savri- l 1968.

[1.8] FIERO, I.B. et al.. Status of erbium burnable absorber development at ABB Combustion Engineering Nuclear Fuel, International Topical Meetin Lighn go t Water Reactor Fuel Performance, West Palm Beach, Florid a, USA , April 1994

16 . 2 UNIRRADIATED GADOLINIA FUEL PROPERTIES

2.1. STRUCTURE AND CHEMISTRY OF THE (U, Gd)O2 SOLID SOLUTIONS

2.1.1. Principal properties of GdO 2 3

Goldschmidt and co-workers recognized three crystallographic structures of the rare- earth oxides [2.1] temperaturs A . e increases cubiC , c (b.c.c. wit molecule6 h1 unie th t o st

cell - TICI2 type) transforms into B monoclinic, and finally into A hexagonal. Shafer and Roy [2.2] indicated that onl C-type yth e structure exist r rarsfo e earths wit atomin ha c number greater tha , than65 t onl A-type yth e exist r lanthansfo ceri d thad B-fore aan a an th t ms i stable for samaria and gadolinia. They indicate the probable existence of the A-type structure for gadolini samarid aan t aelevatea d temperatures existencs It . e probably depende th n so startinpurite th f yo g material duratiod san heaf no t treatment accordin Guentergo t Mozzd tan i [2.3]. Rotd Schneidean h r [2.4 d others]an , (e.g. [2.5]), observeB o t dC thae th t transformation is not truly reversible and occurs at approximately 1250°C or, according to Warshaw and Roy, at approximately 1200°C [2.6]. Thus, it could be concluded that dependin previoue th n go s heat treatment, gadolinia oxide could exist eithe cubiC n i rc form (for heat treatments at temperatures below 1200-1250°C, depending on time of exposure) or in B monoclinic form (for heat treatments at temperatures higher than 1250°C, depending hean o t exposure).

Tabl 1 comparee2. s some characteristic ZrBd :an f Gdlatticso 0UO , e typd ean 22 3

parameters, theoretical density (TD), mean2 linear expansion coefficient for 20-1200°C region (a), elastic Young's modulus (E), shear modulus (G) and Poisson's ratio (v) calculated for zero porosity materials.

Unit cell dimension three th f eso form f Gdso give2e 0 3ar Rot y Schneidend b han r [2.4] as follows:

A-type: a = 3.76 A, c = 5.89 A (hexagonal);

B-type: a = 14.06 A, b = 3.572 A, c = 8.75A, ß = 100'10" (monoclinic);

10.812= C-typea (cubic2A : , b.c.c., TICI2 type).

More detailed informatio elastin no c propertie microcrackind an s f Gdgo d 2Oan 3

compositions Gd0-Hf0 gives i Doly Hunted nb ean r Cased [2.8an ] , Smyt Hunted han r 3 2

[2.11]. 2

2.1.2. Types of (U, Gd)02 solid solution

consideree b n Gdca 0 havins da cubiga c fluorite-type structure with one-quartef o r

2 3 anioe 2 ion" O th sn e i sub-latticth e vacancy nearnese . th Thereforecatioe o t th e f so n du d ,an

radii, solid solution cubif so c fluorite-type structur easile ear y forme U0e th 2 -Gddn i 2O3 system

Zr0e th 2n wel-Gdi s a (s a l 203 system). Charge compensatio UOe th 2n -Gdi 203 systen mca

be realized through vacancy formatio anioe th nn i (oxygen) sublattice, oxidatio4o t U f no

+ 6+ Uo5 r U, interstitial formation or a combination of these defect types, depending on the sintering atmosphere.

This aspec consideres wa t Radford an o detain di H d y [2.12]b l . They classified nine

models r verfo : y reducing (dr atmospher) H y e (models 1-3)r normafo ; bubble(H l d through 2 2 a water bath) atmosphere (model morr sfo 4-5)ed oxidizinan ; g (CO bubble2 /CH r Oo d through a heater water bath) conditions (models 6-9). Formulae, defect types and 0/U and O/M ratios for these models are given in Table 2.2.

17 TABLE 2.1. SOME CHARACTERISTICS OF Gd203, UO2 AND ZrB2

Oxide Lattice Parameters TD Tm a E G V type (A) (9 cm"3) CO (K'1) (kbars) (kbars)

Gd203 monoclinic 14.0a= 6 7.403 2330 1502 588 0.2771 3.57= b 2 [2.5, 2.7] 8.7c= 5 [2.8, 2.9] 100'10ß= " or cubic a = 10.81221

TICI2

UO2 cubic 5.468a= 6 10.96 2860 12.0 2300 874 0.316

[2.10] CaF2

ZrB2 hexagonal, a = 3.17 6.092 3040 7.7 4200 [2.10] AlB2 3.53c= 3

2.1.3. Lattice Parameter

Fonear-stoichiometrie th r c (U,Gd)0 solid solutions, lattice parameter decreases with 2 increas f Gdeo 2O3 conten contend G mol.%0 4 t o leas t f a t o tp u t.

The variatio e latticth n ni e paramete ) wit contend (a rh G , expresseX t mols da e fraction formulae gives i , th y nb :

5.474-0.19= a 0[2.5X ]

(for mechanically mixed Gd fuel; sintered in H2, 1700°C,4h; 0-40%mol.Gd203);

a = 5.470-0.162 X [2.5I

(for mechanically mixe fueld dG ; sintere , 1700°C,4hAr n di ; 0-40%mol.Gd203);

a = 5.4704-0.237 X [2.13]

(for coprecipitate fueld dG ; sintere , 1600°C2 H n di ; 0-3 Gd2h o , 0w/ 2 O3);

5.4725-0.18= a 5[2.14X ]

(for mechanically mixed Gd fuel; sintered in H2, 1800°C, 2h; 0-15 w/o Gd2O3);

a = 5.4725 -0.185 X [2.14])

(for mechanically mixed Gd fuel; sintered in humidified H2, 1700°C, 4h; 0-12 w/o Gd203).

Summarizing these results it is concluded that for near-stoichiometric (U, Gd)O, the 2 solid solution lattice parameter decreases with the increase of Gd203 content (=0.002A per

Gd% O l ) independentlmo 1 e pelleth f to y fabrication route (mechanical mixinr o g 3

2 coprecipitation).

18 TABLE 2.2. O / U AND O / M RATIOS FOR THE DIFFERENT STRUCTURAL MODELS [2.12]

x~y Cation Anion Model Formula Interstices O/U O/M y«z Lattice Lattice

1 . Oxygen vacancy (U'+Gd^O^, 0 y= Full Defect None = 2 <2

G 3+ 2. Cation interstitial (u?.W Oo2 y = 0 Full Full (x/4)Gd = 2 <2

3. Us*and oxygen vacancies y x> Full Defect None >2 <2 (U^x.yUy Gdx )O2.(x-y)/a

+ 5+ 4.U (ut;yu; Gdr)o2 x = y Full Full None >2 = 2

5. U6+and anion interstitial x 2 >2 O) 2+(yx U( .Gd x)/1 .«.2y yU

6+ 6. U (U^2)xUrGdr)02 2 x/ z= Full Full None >2 = 2

6+ + 2 7. U and anion interstitial U?.t-( z Uz Gdx*) O2+z.(x/2) z > x/2 Full Full [z-(x/2)]0 - >2 >2

+ 3+ x = y + 2z 8. U5+an+ e dU Full Full None >2 _ p (ut;y.zuru; G^ )o2 z 2 y=

z 2 + y x> 9. U5+, U'^and oxygen vacancies (Un. . UrU^G^+)0 . . . Full Defect None >2 <2 y z 2 (x y 2z)/2 z 2 y> 0.547-

fc 0-546-

1 2 ce CL S 0.545 *i

0.544-

0 0.01 0.02 0.03 0.04 In U,. Gd X y y O2.y

1 I ! I i

2.00 1.99 1.98 1.97 1.96 O/M ratio

Fig relatione . Th 2.1. between lattice parameter degreee th d oxygenf o an deficiency, or O/M ratio, for Ui.yGdy02-y solid solution. The gradient of parallel straight lines represents dal dx at constant concentrations of gadolinium [2.14].

TABLE 2.3. RELATION BETWEEN O/U RATIO AND Us+ CONCENTRATION [2.12]

5 5+ Gd2O3 u 7u U / U+Gd O/U ratio (mole fraction) (w/o) (mole fraction)

2.01 0.10 6.94 0.02 0.018 2.01 0.20 14.37 0.02 0.016 2.01 0.30 22.34 0.02 0.014

2.03 0.10 6.94 0.06 0.054 2.03 0.20 14.37 0.06 0.048 2.03 0.30 22.34 0.06 0.042

2.05 0.10 6.94 0.10 0.090 2.05 0.20 14.37 0.10 0.080 2.05 0.30 22.34 0.10 0.070

20 mentiones A d stoichiometribefore th r efo , Gd)Oc(U solid solution mixe,a d modef o l

"oxyged "Uan 5+" n vacancy" (Table 2.2) look2 s more suitable whicn ,i (0.88Ä+ 5 hU smalles )i r than U4+ (1.001 A), and the lattice parameter decreases even though relatively large Gd3+ (1.053Â) ionintroducee ar s d (ion radii were taken from [2.13]). This dependencs ewa deduce Ohmichy db . [2.14 al give o t e derivativie wh ]s th f latticeo e paramete wit) d (a rh G content (X) as:

^ - ~

4+ 5+ 3+ For hypostoichiometri , Gd)0c(U 2 solid solution f (Uso 1.2x+2yU x.2vGd x)02.y typt ei shows wa n [2.14] tha lattice tth e parameter increases wit n rati M hdecreasa a r 0/ oo e th en i increase of oxygen vacancy concentration at constant concentration of gadolinium (Fig. 2.1).

Omich [2.14. al t e i ] explained that this increas lattice th e en ei th parameteo t e du s rwa size of the oxygen vacancy, which is 10% larger than that of the O' 2 ion. However, Ho and Radford [2.12] connected this dependence with a fewer number of small U5+ ions being developed as the number of vacancies increases at constant Gd content.

Fopartiae th r l non-equilibriu , Gd)0m(U solid solutions (du insufficieno t e t timd ean

/or temperature of sintering) tw2 o or three types of solid solutions with different Gd content can be observed with markedly different lattice parameters.

ratio oxygeM d O/ s an d 2.1.4nan potentialU .O/ (U,Gd)Oe th sn i solid solutions 2

Diffusion-controlled phenomena of constituent ions in U02 and (U, Gd)02 such as sintering, creep, grain growt fissiod hreleasan s nga e depend principall thein yo r composition [2.14, 2.15]. Thi becauss si e these phenomen rate-controllee aar slowee th y db r cation diffusivity.

Theoretically, the O/U ratio of the solid solution is controlled only by the number of uranium cations in valency states other than U4*. Proceeding from this assumption Ho and Radfor, Gd)O(U e ratiosoliM dth r O/ estimated sfo d solutionan U dO/ s (see Table 2.3d an )

relationshie th 2 Ue pth /betweed U ratian U on O/ [2.12] . 5+ Their results showed tha smallee tth Ue 5+rth /U ratio closee rati,th U 2.00o rot O/ e 0th will be. Thus, in order to obtain the optimum properties of Gd fuel, it is essential to minimize + ions6 U .d adjuste e b Althougan ratiamoune U n + th 5 oO/ sinterinca y U db e f o htth a n gi 5+ 6+ controlled CO2 atmosphere, this has the effect of increasing the U and U content. Ho and Radford proposed that the addition of a doping agent such as Nb2O5 would be more beneficial [2.12].

The relation between oxygen potentialAG02 and O/M ratio was measured by a thermogravimetric (TGA) technique for U02+x and (U^Gd^O^, where 0.02 < x < 0.08 and 0.04= y , 0.1 0.2d 4an 7 mole fractions [2.16, 2.17]. Samples were obtaine mechanicay db l

blending, pressin atmospher2 sinterind H gan n a n gi t 1700°. ea Result2h r C fo f thes so e measurement t 100sa 1300°d 0 an gived 2.3e Can ar Fign i 2 . s2.

The stable hypostoichiometric phase is seen for the U-l.yGdyO2 solid solutions at oxygen potentials above -400 kJ/mol, even though such a phase is not seen for U0. The

stabilization hypostoichiometritrene th dn i c UGd. 0 fue explaines i l increasn a e y dth b n ei 2 y y 2

oxidation state of the uranium ions1 , thereby producing more positive oxygen potentials.

21 -200 -

Ä -300 -

«4 |oO

-400 -

1.98 2.00 2.02 2.04 2.06

O / M ratio

Fig. 2.2. Oxygen potentials r U0fo .ggGd 0.o402+x plotted functionratioa M I s a 0 f o at 1000 °C [2.17]

1300= CC

-200 -

o .E 2, -300

-400

1.98 2.00 2.02 2.04 2.06

O / M ratio

Fig. 2.3. Oxygen potentials for UOM GdOM 02+x plotted as a function of Of M ratio at 1300 °C [2.17]

22 -200

-300

O .E -400 -

CM 'o <3

-500

yOz 000±0.001

UO2 -600 - Blackburn Model [2.18]

J_____L 0 80 1000 1200 1400 1600 Temperature (°C)

Fig. 2.4. Oxygen potentials for exactly stoichiometric U-\^Gdy O2 solid solutions [2.17]

oxygee Th n potential near-stoichiometrir sfo c (O/ 2.000JL0.001= M . U, d an 2 UO ) yGdyO2 solid solution givee s ar Figs n i . 2.2.; 2.32.4d .an . [17] solie .Th d circles represent the AG02 values at O/M = 2.000. The hatched area represents the AG02 values within the O/M region from 1.999 to 2.001. The values of pure stoichiometric composition, calculated fro Blackbure mth loweJ k n0 rmode 14 tha e o t n th l 0 [18]9 e ar , data obtained by K.Une et al. [16, 17]

These results suggest that further experimental work is needed to obtain more data temperaturw lo e botth hn i e range (1000° hige C - 1500°Cth h n i rang d an )e (1600°C- 1100°C) in order to improve the oxygen potential assessment of the (U^Cdy) O2.x solid solution for low Gd content fuel (y < = 0.1).

2.2. Physical properties

2.2.1. Theoretica sintered an l d densit Gd}0, (U f y2o system

Calculations of the theoretical (or X-ray) density (TD) of the (U, Gd)O2 solid solution require knowledg chemicae th f eo l composition, lattice paramete structurad an r l model.

23 For example, the density of a solid solution with X mole fraction of Gd203 (adopting the "oxygen vacancy" model item 1 in table 2.1.) is given by

- 4 2380-*) + 157» + 16 (2-jffl (1) 0.602 3 103a 24

Anion exchange EDTe th , A titration method [19 spectro-photometrid ]an c methods usuall[20e ar ] ydeterminatioe useth r dfo contend G Uf d no 6+ an t/U 4+ ratioe Th . O/(U + Gd) ratios can be calculated using the following equation

Wr. 6 + 4 (2) M, LGà

weighwhera s i atomin ea W t s fractioi cM weightd nan .

Ho and Radford [12] used the experimental data (Gd content, U6+/U4+ ratios, O/(U + Gd) ratios, lattice parameters) obtained by Fukushima et al.[21] for coprecipitated

UO2-Gd203 solid solutions wit contentmol5 d 1 hcalculateG d %o t an usin D p su T ge dth different structural models (Fig. 2.5.). The line plotted above Fukushima's is from the "U5+" model showin lowee th gd r highelinan , froes i "oxyge e mTD r th n vacancy" model.

The analysis of Fukushima et al. implied that U5"1" ions were more likely than U6+ ions, especially taking into account that the pellets were sintered in an Ar/H2 mixture [21]. consideres i t I d tha adequatn a t e representatio fued G l pellete expectee th th f f o no s D dT is attained when using the models developed by Ho and Radford.

11.0

10.9

10.8 Oxygen ^ Fukushima et al. u Vacancy 10 7 Model1 - U*+Model 4 Q 10.6

10.5 U"and Anion Interstices Model 7 10.4 J_____I_____I_____I

2 4 6 8 10 12 14 16 18

Gd2O3 content (wt %)

Fig. 2.5 3 densities.O 2 ComparisonGd - 2 databasede UO Fukushimaf th f o n o o . [2.21]al t e in relation theoreticalto models12] [2.

24 100-

99- •

98--

97- •

96- •

94--

93- •

92- • Westinghouse 91-- data

90- 45678 10 11 12 Gadolinia content, w/o

Blended only B&W data

90 3456782 1 9 1 1 10 Gadolinia content, w/o

Fig. 2.6. Effect of gadolinia content on sintered density [2.12, 2.13].

25 Three equations describing the TD are listed below:

p = 10.962-0.0348x [2.21]

10.96p= 0 -0.033X [2.13]

10.96p= 00.02- x0.0027x- 0.00015x+ [2.22] 2 3

3 where p is the TD in g/cm and x is the Gd203 content (w/o), valid for 0 < x < 12.

It was shown that up to 10 w/o of Gd203 content, the quantity of the U02 phase in a (U,Gd)02 solid solutio significano n s n ha [2.21] D T t e effeccomparison I .th n o t n wite hth diminishin witD gT h increasing Gd 2dependencOe 3th f sintereeo d densit percentaga s y(a f eo mors i ) e TD complicatede th simple th r efo , reason tha solutioe th t naturns it typd n ean ca

undergo changes with increasing Gd2O3 content.

6 show2. Fige effec. sth f Gdo t 203 contensinteree th n o t d densit f materiao y s a l observe differeno tw y db t sources [2.12, 2.23] Westinghouse Th . e specimens o (datH y ab

and RadfordW ) & wer t B 1750°C a detaileo 2 e n H sintere th t t f x hourso bu ,si we r n si dfo sintering condition givene sar . Similar data wer Ogumd e an obtaine e a Un [2.17 y db ] where

samples were sintere flowin a atmospher2 dn H i y gdr hour2 seer e t e1700°Cb fo sa nn ca t I . that ther considerabls ei e disagreemen effece th sintereo e t t th s tha contena d tn o G t ds ha t density.

It has been shown that Gd reduces cation diffusivities and sinterability, especially for

a dry H2 sintering atmosphere. Sintering in a wet H2 atmosphere could result in enhanced oxidatiosmalleo t correspondinglion+ d 6 + 4 U s an U d rf nUo an 5"1" y quicker cation diffusion (for

4-8 w/o Gd203, Ho and Radford [2,12]). In the case of the B & W experiments, increased sinterabilit achieves ywa implementiny db gpowdea r milling process (Newma l [2.23])a t ne .

2.2.2. Thermal conductivity

Of all the properties of Gd fuel, thermal conductivity has probably received the most attention over the past 10 years. This is because as gadolinia is added to UO2 its thermal conductivity reduces significantly, which, reactorn i , lead higheo st r fuel temperature l othesal r conditions being equal.

However, the experimenters disagree on the magnitude of the reduction in thermal conductivit shows7 2. Figs ya . . This presents results from four different out-of-pile sources [2.21, 2.22, 2.24, 2.25] and shows the ratio of the thermal conductivity of gadolinia-doped

UO2 to undoped UO2 for various gadolinia contents (w/o) as a function of temperature. All show thaeffece th t f gadolinio t reduco t s athermai e eth l conductivit thad yan t this effect becomes less important at higher temperatures, but the graphs show marked difference in absolute levels.

No detail measuremene th f so t technique sExxoe useth r ndfo dat availablee aar t bu , the other experiments used the laser-flash technique to determine the thermal diffusivities of samples. This technique involves applyin shorga t duration puls facf energe a eo f on eo n yo sampl measurind ean temperature gth opposite th ris n eo e functiofaca s ea f timeno e Th . thermal conductivity is then calculated from the thermal diffusivity, specific heat capacity and density of the samples, the density being corrected for thermal expansion in most cases. The French experimenters also performe seconda d method involvin direce gth t measuremenf o t thermal conductivit steady yb y state radial heat flow which showed reasonable agreement with their laser-flash results.

26 11 11

l - - f •»•

8 - •

- - 7 u

• s- Exxon data

0 200 400 600 800 1000 1200 1400 160 0 010080 1800 60 1200 0 200 14040 0 0 20 160 00 1800 20OO Température 'C Temperature. 'C

^ 9 — • 4 14 w/o 099 w/o

8 -•

- - 8 06 w/o

6 -- 6 _.

2 1 o w/ 2 0 . _ g CEA-Fragema S - • JAERI data o w/ 13 03 data

0 200 400 600 800 1000 1200 1400 1600 1800 2000 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Temperature, *C Temperature, C

Fig. 2.7. Thermal conductivity measurements functiona s a f o gadolinia content temperatureand [2.21, 2.22, 2.24, 2.25].

Individual experimenter's calculations were as follows:

a. B & W data [2.24]

The expression used to calculate thermal conductivity was

k = ad CPp0 (1 + 3a9)-1

wher - therma : d ea l diffusivity

CP: - specific heat capacity

P0: - room temperature density

ae: - thermal expansion coefficient

27 The specific heat was derived from measured enthalpy data using a Bunsen-type ice- drop calorimeter. Thermal expansion was measured by a recording quartz dilatometer over rang e opticay 900°o th t b ed O Can l dilatometry ove2000°C o rangt e th r0 therma e e90 Th . l conductivity data were correlated for each gadolinia concentration as a function of temperature T(K) using expression forme th f s:o

ET)'+ A 1( = k

where A and B are constants normalized to 95% theoretical density. Values for A and t givenno e B.ar

b. Exxon data [2.25]

No detail availablee sar .

c. CEA-FRAGEMA data [2.22]

The expression use calculatdto e thermal conductivitywas

p p C à a = k

where ad: - thermal diffusivity

CP - :specifi c heat capacity

p: - density

Specific heat capacity was derived from enthalpy measurements using differential microcalorimetry. It is not clear whether thermal expansion effects were considered but if

they were, pur valueeU0 s would have been used. 2

No correlatioe detailth n o s f thermano l conductivity data with temperature ear provided, althoug BT)'+ A ( relationshin ha agaips i n likely. 1

d. JAERI data [2.21]

The expression used to calculate thermal conductivity was

Cd a pp = k

wher therma- : d ea l diffusivity

CP specifi- : c heat capacity

p: - density

Specific heat capacit estimates ywa d from weightea d averag f previousleo y published data for specific heats of U0 and gadolinia, which had, in any case, to be extrapolated to

temperatures outside the range2 s for which they were originally intended. Thermal expansion

effects were include densite th n di y term t valuebu , r pur2 wersfo eU0 e used. The thermal conductivity data were correlated for each gadolinia concentration as a functio f temperaturno arouno t p u dK e1300°T C using expression fore th m f so

BT)-+ A 1( = k

TABLE 2.4. VALUES OF CONSTANTS A AND B IN THE EXPRESSION FOR THERMAL CONDUCTIVITY k=(A + B T)' 1 AS DETERMINED BYJAERI [2.21]

Gadolinia content A B (w/o) (10'2mK/W) (10-2 m/W)

0.00 3.29 0.0236 0.99 6.53 0.0235 2.03 10.18 0.0233 4.04 13.37 0.0234 6.03 17.29 0.0229 8.10 22.41 0.0220 10.33 27.19 0.0210

Not»: temperaturesK mustT in be

where A and B are constants normalized to 95% theoretical density. Values for A and B are give Tabln i e 2.4.

e. ARSRIIM and RCC Kl (Russian Federation) data

The result f calculationso valueindireck n a f so y s(b t measurement method r nearfo ) -

stoichiometric (U, Gd)O pellets with Gd0 contents of 5 and 6 w/o [2.26] are similar to 2 3

JAERI (Fukushim l [2.21]a t e a ) data2 .

In assessing these data, it is clear that apparently similar experimental techniques have yielded dissimilar results. More credibility could be attached to the B&W results, since all parameters calculatioe useth n di f thermano l conductivity were obtained fro samplee mth s tese useth t n di program JAERe Th . I results used extrapolated hybrid specific heat data employin rule mixturesf gth eo . JAER FRAGEMd an I A used thermal expansion data pertaining

to pure U02, but the effect of this should be negligible. There are no details available for the Exxon data.

2.2.3. Specific heat capacity

This is an important characteristic, especially as it is used by most experimenters to

derive thermal conductivity simplicity'r Fo . s sake specifie th , c heat capacits i f 2 puryo eUO often use r gadolinia-dopedfo Figd 8 showan 2. . 2 sdU0 that reasonabla thi s si y good approximatio mosr nfo t doping level temperatureo t p su f arounso d 700°C. Fig.2.8 presents

specific heat capacity dat gadolinia-doper afo 2 fro differeno dmU0 tw t sources [2.22, 2.27] and compares each with standard data for UO2. The heat capacity is seen to be weakly dependen dopinn o t g arouno levet p u l d 700°C Japaneso (witw/ exceptioe 0 h1 th ee th f no data) e maximuth , m deviation fro 2 date purmth aUO e being aroun . Abov5% d e this temperature, however, Japanese data show a significant increase over the pure U02 data, whic temperaturhs i contend G d ean t dependent.

29 500- 500-

10 w/o I $ 400--

D. o. 0 w/o oO oCO 0 w/o —J o<0 X 8 w/o o •I5 300- -= 300-

CEA-Fragema Japanese data data

200- 0 200 400 600 800 1000 1200 140 00 16010080 00 60 180120 0 040 0 2000 14020 0 0 0 1600 1800 2000 TemperatureC * , Température, 'C

Fig. 2.8. Specific heat capacity measurements as a function of gadolinia content and temperature [2.22,2.27].

Individual experimenter's calculations wer followss ea :

. a CEA-FRAGEMA data [2.22]

Specific heat capacit obtaines ywa differentiay db l microcalorimetr measuriny yb e gth enthalpy variation during thermal cycling of the samples. The expression used to calculate specific heat capacity was:

Cp = 337.7 - (1.823-10-2T) - (0.997-10'7 T);

where T is temperature in K and CP is given in J/kg K. Data were obtained for 8 w/o gadolinia only.

. b Japanese data [2.27]

Specific heat capacity data were obtained for a number of different Gd content levels

of the form UGd. 0 up to 1 500 K by means of direct heating pulse calorimetry. The data y v 2

were fitted to 1 equations of the form

3 2 1 p dC + ET + DT + CT + T B + A C= p

constante ar E d san giveD , C wherTabl n i , B , etemperatur, s ei A 2.5 dC T , d an K en i the excess heat capacity, is given by

2 2 { (AH)dC = p(AS/R p (-AH/RTex /RTp } ex ) )

enthalpe th e entropd ar yan S A f activatio yo d wheran H eA n respectivelye th s i R ,

universal gas constant 8.314 J/mol-K. and Cp is given in J/mol-K.

It is clear that both experimenters have found that the specific heat capacity of

gadolinia-U0 ver s contenti d yG arouno y clost p an thau ,o r e t df purfo o t 700°C eU0 e Th . 2 2

30 TABLE 2.5. VALUES OF CONSTANTS A, B, C, D AND E, AND ENTHALPY AND ENTROPY OF ACTIVATION (JH, 4S) IN THE EXPRESSION FOR SPECIFIC HEAT CAPACITY AS DETERMINED JAPANESEBY [2.27].

Gadolinia A B C D E //H ^S content IQ'2 104 107 109 103 (w/o) (J / mol) KK 2l ) mo ( J/ (J / mol) (JK/mol) mol/ 2 (JK ) (J mol/ ) (J / mol K)

3 45.905 1.7805 3.5171 -1.6717 2.1997 106.6 29.5

5 55.330 1.4582 2.3691 -1.1677 1.4657 90.9 28.9

7 61.215 1.2100 1.9362 -1.0336 1.3170 71.2 22.1

10 38.596 2.2130 4.0941 -1.7856 2.2220 53.3 21.2

Note: temperaturesK mustn i T e b

data beyond this temperature shows a clear deviation from the standard U02 data and from the values that the rule of mixtures would predict. Inaba et al. [2.27] postulates the formation of oxygen clusters as being the reason for the excess heat capacity, but experimental error effects and scatter of results could equally account for a large part of it. It is clear that further investigations at these higher temperatures are desirable. Also, where experimenters have used pure-UO2 values for specific heat capacity, this may have unduly pessimised their calculations of thermal conductivity.

2.2.4. Thermal diffusivity

Thermal diffusivit , Gd)O(U f yo 2 , from which thermal conductivit usualls yi y derived, anothes i r important propert estimatinr yfo fuee gth l behaviour.

Generally factore ,th s affectin measuree gth d value f thermaso l diffusivit roughle yar y classified into chemical factors suc displaces ha d Gd3+ ionconsequentld san y oxidized U5+, and geometrical factors such as microcracks, pores and cracks.

TABLE 2.6. THERMAL DIFFUSIVITY SAMPLE CHARACTERISTICS

Reference Gd2O3 content Fabrication Sample Density Presencf eo route thickness microcracks (w/o) (mm) (% TD)

mechanical blend severan i l [2.28] 0, 10 0.5 - 1 .2 93.6 - 98.5 and co-precipitation samples

[2.26] 0, 6, 8, 10 mechanical blend9 0. - 8 0. ~ 95 no

31 [2.28]

1.5 A-6;O-8;O-10wt%GdîOa [2.26] —— -10 wfitGcfeOa- without microcracks [2.28]

-10 wt% Gd2 03 - with microcracks [2.28]

.to - -- -10wt%GdzO3 [2.21] CM E,

Î..O to

I 0.5 -

500 600 700 800 900 10001100 1200 1300 1400 Temperature (°C)

Fig. 2.9. Thermal Dfffusivity of UO2 and (U, Gd) O2 samples (normalized to 95 % TD)

The therma (U,Gd)0l d diffusivitan 2 U0 2 f sampleyo measures s wa lase e th ry db puls e method ove temperatura r e rang f 473-127eo [2.263K 300-200d ]an [2.28]0K . Principal characteristic f sampleso give e measurementd sar Tabl an n i 6 e2. s result Fig.2.9sn i . Data obtaine Fukushimy db . [2.21al alst e ae oar ] give Fig.2.9n i .

The principal results of the thermal diffusivity measurements [2.21,2.26,2.28] can be summarize followss da :

the data are independent of sample thickness and fabrication route;

thermal diffusivity of (U,Gd)O fuel is lower than that of UO fuel at low 2

(< 1200 K) temperatures, the2 magnitude of the difference depending on Gd concentration;

at higher temperatures (near 2000 K), the diffusivities of the (U,Gd)0 solid 2 solution and U02 are almost the same;

below 1200 K the diffusivities of the (U,Gd)02 samples with microcracks are about 30% lower than the microcrack-free pellets and agree with the data of Fukushim . [2.21]al t ae .

32 2.2.5. Thermal expansion

Wada et al. [2.13] measured the thermal expansion coefficient of Gd fuel

(coprecipitated, sintered in H2 atmosphere at 1600°C, 24h) in thé composition range 0-30 w/o

temperatura o t Gp du f 122eo 3usinK dilatometerga . They found tha additioe tth f Gdno 0

3 2 1

up to 12 w/o in UO2 had little effect on the thermal expansion coefficient ( = lO.S-IO^K' ).

Beyond 12 w/o, it increased slightly with Gd0 content to 11.7-10"K6 at 30 w/o GdO. 2 3 2 3

Chotard et al. [2.22] (for mechanically blended samples, sintered in a humidified H

atmosphere at 1700°C, 4h) obtained results consistent with those of Wada et al. [2.13].2 They showed that ove temperature th r concentratioa e d 200 o rangt an 0 0K e 30 n rango t e0

Gdo 12w/ 2 Othermae 3th , l expansion coefficien r pursame fo th es s e a i tUO 2.

Une [2.29] measured the linear thermal expansion of Gd fuel (mechanically blended, sintere flowina dn i gatmospherH t 172ea 1750°Cr 0o compositioe th n i ) n rang f 0,5,8,1eo 0

w/o Gd0 ove2 r the temperature range of 298 to 1973 K using a dilatometer. Results indicate 2 3 tha lineae tth r thermal expansion coefficient becomes larger with increasing Gd2O3 content. At 1973 K the linear thermal expansions for U02- 5 w/o Gd2O3/ U02 - 8 w/o Gd2O3 and U02 -

o Gdw/ 0 wer0 1 e abou 6.2d t 2.7an % 4 ,large5. r tha pelletse valueUO nth r fo s, 3 2

2 respectively.

The linear thermal expansio expressee b followine n th nca y db g least-square fitted equation:

2 CT+ T dL/LB ; + 0=A

where T is the temperature in K and L0 is the specimen length at 298 K. Table 2.7 gives the regressio temperature th r n fo obtaine constantC e d Un ean y o 197t db rangB s8 A, 3e29 K [2.29].

B & W data by Newman [2.23] were obtained by quartz and optical dilatometry for

(U,Gd)0 pellets with GdO contents of O, 5.66 and 8.50 w/o (Fig. 2.10). They are similar 2 3

to the dat2 a of Chotard [2.22] and Une [2.29].

Summarizing all the above-mentioned results, it is considered that the mean coefficient

of linear thermal expansion shows negligible variation from values for pure UO2.

TABLE 2.7. REGRESSION CONSTANTS FOR EQUATION OF LINEAR THERMAL EXPANSION OF 2 UO2 -Gd) [2.29]T zC O 3+ FUELT B + (dL/LA = 0

Gd2O3 content A B C Standard deviation 10* IQ"6 IQ'9 10* (w/o) 0 -2.238 7.165 2.095 5.2 5 -2.314 7.358 2.156 7.1 8 -2.391 7.433 2.278 9.4 10 -2.284 7.162 2.430 7.4

33 c oto.

B&W data

0 10080 00 60 120 0 040 140 0 20 0 1600 0 1800 2000 Temperature, 'C

Fig. 2.10. Linear thermal expansion measurements as a function of gadolinia content and temperature [2.23].

2.2.6. UO2 - Gd2O3 phase diagram and melting temperature

factoro Tw s shoul takee db n into account before - Gd analysin 2 2OU0 3 phase gth e diagram:

1 ) Solid solutions of the rare earth oxides and urania form readily. Materials whose

ionie ionith c cf e rar o radith withiradie e f ar % io earth i n20 s form solid

+ solutions [2.30]. The ionic radii of urania and gadolinia cations are: U0.94 7 A, U6+ 0.8 A and Gd3+0.97 A.

) 2 Rare earth oxide e stabilizerar s f n uranioxidizino sa n i ad reducin an g g environment [2.15].

Phase diagrams of (U,Gd)O solid solutions obtained by two different methods

(mechanical blending [2.15 coprecipitatiod an ] 2 n [2.13] showe ar ) Fign ni . 2.1 2.1d 1an 2 respectively. X-ray diffraction analysis indicate botdn i h case presence sth .c.cf f onlee o .yon solid solutio t gadolinina a content 30-4o t p su 0 mol. lineaa d %ran decreas solie th df eo solution lattice parameter with increasing gadolinia content.

Specimens (pieces of crushed pellets) were put directly on a tungsten (W) heater and heate iner atmosphern s a tga dn i 5%H+ r [2.15e A [2.13])ed (H ]an . Liquidus temperatures

represen complete th t 2 e meltin compositee th f go .

Liquidus temperatures from both studies decrease practically linearly with increasing gadolinia content (from 2800 to 2760°C [2.5] or 2800 to 2700°C [2.13] for Gd content up w/o)0 3 o t . However solidue th , s line, reporte t 2250°C a Béai. [2.15y de b al b t s se o t ]wa , t founno Wad y . [2.13]db al t ae . Liquid phases sometimes observe t graida n boundaries were identified [2.13 W-U-Gd-s a ] r Mo-U-GOo d (whe nmolybdenuma , heate useds Mo , t wa r bu ) no U-Gd-0s a t . Therefor assumes wa et i d [2.13] tha solidue tth s line would probabl vere yli y liquidue closth tha d o ephase t sth tan e relationshi (U,Gd)0e th f po 2 system coul similae db r

to that of the (U,Gd)0 system reported by Grossman et al. [2.31] and Chotard et al [2.22]. 2

34 Temperature (°C) Temperature (°C) ro ro o r to o r 10 ro 8

oI o o ,.&' o \ o o S f CD o o <5 Q • Öl——l &> \ / W 10 ssgi M U IO Y I •o Q 01 o Q. ff K> O u S7> l £g n O * ^ o o o

I T3 M Öl

o o o

u cn

Ul 8 i i i • Melted 3000 * Partially Melted • • No Change

2900 - • • Liquidus A A Y , — ü ^ ^ ^—^1 1 ,„1 O "''•-^TI"""— > i— A. A T 2800 i_ " r^-x i ~~^T +~~ A T " ^' * 1^ "^ * >. / **«. 1^ ^ ^x Solidus %"-. | 2700 *"" X ^--^^x ^^ X -^ X N 2600 Solidus ^^ • x • Grossman et al. [2.31] \ \ \ 2500 " I Uquidus Wada et al. [2.13] \ " i i 0 5 10

Gd2O3 content (wt %)

2 phaseFigO ) . 2.13Gd , diagram(U . [2.32]

The left corner of the (U,Gd)0 phase diagram was recently investigated by Matsuoka

[2.32]. GdO content pellete w/o0 pellete th 1 f sd Th o .s san wer wer6 e0. e heateo t p du 2 3

the specifie2 d temperature in an Ar gas atmosphere, held there for a specified duration and then cooled rapidly. The specified temperature was selected at 20-30°C intervals around 2800°C liquidue Th .solidu d san s were determine visuay db ceramographid an l c observation heatee oth f d specimens.

The results of [2.32] are shown in Fig. 2.13 together with those presented by Wada [2.13 Grossmad an ] n [2.31 r comparisonfo ] thesn I . e experiments solidue th , e sb linn eca

observed separately from the liquidus for 10 w/o Gd203 pellets, the solidus temperature being 2700-2725jt18°C.

Summarizing the above data, it is concluded that the solidus line lies close to the liquidus althoug difficuls i ht i specifo tt exace yth t temperature difference between these lines.

Although many reports have been published concernin meltine gth g poin f (U,Gd)0o t 2, meltine th g finallt poinye stilt s ti y no lestablished principalle ,du instabilite th o yt samplef yo s over 2800°difficulte th d Cf measurinan y o temperature gth t sucea hhiga h level. Daty ab Beals [2.5], Wada [2.13], Jamanouchi [2.33], Matsuoka [2.32] and Hanson [2.34] showed a slight decrease in melting temperature with increasing gadolinia content (Figs. 2.11-13). In contrast with these results Chotard [2.22] showed (Fig. 2.14) tha meltine tth g temperature s largelwa y independen f gadolinio t o (pelletw/ a2 1 s conten wero t p eu t obtainey b d mechanical milling, blending, granulation and sintering in a humidified H atmosphere at

1700° hours)4 r Cfo . 2

36 3000

CEA-Fragema data Toshiba data Joint Japanese data

2900 -

3 ICO f«i0

2800—

2700- 0 l 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Gadolinia contento w/ ,

Fig. 2.14. Melting temperature as a function ofgadolinia content [2.13, 2.22, 2.35].

In a recent joint study of five Japanese utilities [2.35] the melting temperature of fragments of pellets obtained by a master blending technique and sintered in cracked ammonium with stea t m1700°a hour4 studies r Cfo swa d usin tungstega n ribbon heater method. The results of this study are also shown in Fig. 2.14 for two groups of specimens (the quantitwhil% 5 phas U0 es f f yo s tha20-25%)typewa o f typ o wa 1 et e 2 Th .

maximu2 m differenc meltine th en i g temperatur confirmes wa t 40°i s t eCwa d bu that thers ewa

no significant decrease of melting temperature up to 8 w/o Gd0. From 10 to 15 w/o Gd0, 3 2 3

a 50-60°C decreas observes chango ewa n observet s Gdo dbu ewa O w/ . 5 2 d o frot 5 m1 2 2 3 same Th e dependenc meltinf eo g temperatur additivn eo e concentratio obtaines nwa UOr dfo 2- Sm203 and UO2-Dy203 systems.

37 2.2.7. Grain growth

Grain growth behaviour at 1700-2000°C of U0 and (U,Gd)0 fuel pellets prepared by 2

the coprecipitation method and mechanical blending wa2 s investigated in [2.36]. The contents

of Gd203 in the (U,Gd)02 pellets prepared by mechanical blending were 3,5,8 and 10 w/o and thos coprecipitatioy eb w/o0 1 n d wer. an e5

Powder of UO2 and (U,Gd)O2 was pressed at 340-390 MPa, and the green pellets were sintered at 1600°C in hydrogen for 2 hours (95-97% TD). As expected the coprecipitation method provided better homogeneit (U,Gd)0e th f yo 2 solid solution tha e mechanicanth l

blending, where small amounts of free U0 and GdO were observed. 2 2 3

preparee Th d specimens were heate reducina dn i g atmospher f He-8%Heo t 17002a - 2000°C for 1-30 hours with no significant change in 0/M ratio. Grain diameters were measure lineae th y rdb intercep differencee th t o methodt e date du sd aTh . G scatte e th n i r in pellets between mechanical blending and coprecipitation was not too large. Data were best fitte fourty db h power rate equations:

4 14 8 4 3.7= 0 910x D - fo r exp(-142,000/RTD UO2: t )(//m ),

for (U,Gd)O: D - D= 4.98 x 10 expH 4O,000/RT)t (//m)

2 0

4 4 7 1 4 constans ga e wherth t s i (1.98 eR 7 cal/mo temperatur- T , K) l tim- e t (K)ed (hours)an , .

It is clear that the grain growth rate of the (U,Gd)02 pellet is approximately one order of magnitude smaller that the UO2 growth rate.

authore Th f [2.36so ] explaine difference dth grain ei n growth rate betweed an nUO 2 (U,Gd)02 through suppressio atoU f mno diffusio (U,Gd)Oe th n i 2 pellets. They explainee dth absence of this difference for pellets with different Gd contents through keeping the O/M ratio well below 2.0 and an unchanged vacancy concentration.

Littlechil . [2.37al t de ] also observed thaadditioe th t f smalno l amount f gadoliniso a (0-6 w/o) cause decreasda grain ei n size t that gadolinibu ,s ,a a conten increases wa t d above 6 w/o, grain size increased again.

2.2.8U .> - Diffusio d G d nan d coefficienG > - U f to

date aTh described above sho fuewd thaG l r sinteretfo typicaa dn i l atmosphere, cation

diffusio slowes ni r tha purn i e U02.

The kinetics of cation interdiffusion was studied with the help of alpha and neutron autoradiography [2.38] and SEM and EDX techniques [2.39] (see App.V). The diffusion couples were prepare placiny db g Gd203 green pellets powder2 intoUO compactind ,an d gan sintering in a hydrogen atmosphere in the temperature range of 1600 to 1900° C for 4, 16 and 64 hours [2.38] and at 1700 and 1800°C for 100 hours [2.39] respectively. The temperature dependence of cation interdiffusions is given by the equations

D = 3.3 • 10-6 exp (-200000/RT) cm2/s [2.38]

D = 1.0 • 10'5 exp (-260000/RT) cm2/s [2.39]

constans ga e wherth t s i (8,31 eR 4 J/mol-Ktemperatur- T d an ) e (K).

38 Numerical values for typical sintering temperature (1750°C) are:

2.26 • TO'11 cm2/s [2.38]

1.93 • 10'12cm2/s [2.39]

Based on these results, it was estimated that the distance between Gd203 grains should be les 15//~ [2.39sm // d tham 5 an ][2.38n 4- t powdea ] r blendin orden gi obtaio t r na homogeneous pellet by sintering at 1750°C for 4 hours.

2.2.9. Electrical conductivity

Onl investigatioe yon electricaf no l conductivit gadolinia-dopef yo bees ha n foundUO d

[2.40]. Samples with gadolinia contents ranging 2from 2 to 10 w/o have been investigated usin e four-probgth techniquC D e avoio et hige dth h resistance proble contacte th t ma s between probe and sample. Data were obtained during both the heating and cooling process, and no real differences were noticed between the two sets of data. Fig. 2.15 shows the

T (K) 0 10080 0 0 50 0 60 400

:A Up.69oGdo.i4o( )i.99O e 5 d0.1141.99 )O

• ( U 0.914ÖCI 0.06«)O 2.000

o G

1.5 2.5 103/T

Fig. 2.15. Electrical conductivity nearof stoichiometric (U,Gd)0 solid solutions 2 plottedversusJ a s a 7/7. Open solidd an datae marksth e measuredar on the heating and cooling process, respectively [2.40].

39 producresulte fore th th f m o f electrican s i o t l conductivit absolutd an ya e temperature, T , plotte electricae d againsTh . /T l 1 tconductivit sees increasyo nwa t e with temperaturl al r efo samples and to increase with gadolinia content. The values of electrical conductivity of the gadolinia-doped samples were threr o abou o e tw ordert f magnitudso e higher than thosf eo pure U02.

3 MECHANICA2. L PROPERTIES

2.3.1 Elastic modulus

The dat elastin ao c constant limitee sar publicatiode onlon o yt n [2.41] thin I .s paper

elastie th TiO- 2 c 2 constant - NbU0 pellet Gd- 2 d 2 202U0 5an 0 sUO , 3 f werso e measuret da room temperature using the pulse echo overlap method to determine the additive content dependency.

Specimens of U02 with Gd203 content ranging from 0 to 20 w/o, with Nb205 content Ti0d 2an contenrangino w/ 8 gt0. werranginfroo o t mw/ e0 2 prepareg0. froo t m0 d using the mechanical blending method and the coprecipitation method. The specimens were of cylindrica diametelm configuratiom 7-15 d 1 lengthran ~ m 3m 0 1 ,f n boto h end f whicso h were cut by a slicing machine to generate a high degree of parallelism. The natual frequency of the oscillator was 5 MHz.

Fig. 2.16 present result e Young'e sth th f so s modulus measurement Gd- 2 20U0 3 r sfo pellets. The data were normalized to 100% TD using the density correction expression obtained fro experimene mth presentee ar d an t d relativ Young'e th eo t s modulu f purso e UO2. The Gd203 content dependency expressions of the Young's modulus and Poisson's ratio of U02 - Gd203 pellets at room temperature were obtained by fitting a linear regression expression to the data as follows:

3 -5.686.10-1 E/E = 0 CGd

3 6.59.10'- 1 v/v = 0 CGd

Young'e ar 0 v wherd s moduluan Poisson'0 e eE th d san s rati f puroo eGds i UOd 2G 02C 3, content (w/o). 2.0

T3 O

I iCD cCcD

0.0 5 1 0 1 5 20

Gd2O3 content (wt %)

Fig. 2.16. Young's modulus of Gd2O3- U02 pellet vs. GcfcQ>3 content [2.41]

40 TABL£ 2.8. PARAMETERS OF THERMAL CREEP TESTS OF ( U, Gd) O2 PELLETS

Reference Gd2O3 content a t (w/o) (MPa) (°C)

Peehsetal [2.42] 0;6.5 100 1500 Hiraietal [2.43] 0;3;5 8-75 120 1520- 0 Watarumietal [2.44] 3-15 30-60 1300 - 1500

Godinetal [2.45] 0-10 10-50 1200-1400

The Young's modulus and Poisson's ratio of U02 - Nb20s and UO2 - Ti02 pellets remained constant within the range of the experiment.

2.3.2. Creep

There are only a few publications describing thermal creep of Gd pellets [2.42-2.45]. l caseal n I s pellets were obtaine mechanicaa y db l blending method followe sinteriny db g under conditions providing a solid solution. A rated load compression method in vacuum

[2.42, 2.44, 2.45] or in N - 8% H atmosphere [2.43] was used to obtain the data. Gd0 2 2 3

content, compression stresse 2 tesd stan temperature givee sar Tabl n i e 2.8.

Some resultabove-mentionee th f so d thermal cree , Gd)0p(U tes r pelletfo t e sar

presente 2.1g Fi 7dn i [2.45 2.1d ]an 8 [2.43] difference Th . absoluten i e value f creeso p rat2 e between [2.43] and [2.45] can be explained by a significant difference in stress values (12

respectively)a MP 0 3 d stronA an . g dependenc (U,Gd)Od an UO f esamplo e creep stress 2

confirmes wa [2.45]dn i botn I .2 h cases sample densit graid -95s an n y wa sizD % T e ~10//m.

These studie wels s[2.43,2.44a s a l ] concluded that:

1. There is a tendency for a slight decrease of (U, Gd)O2 pellet creep rate with increasing Gd203 content. This agrees with other cation intergranular diffusion rate-determining phenomena such as grain growth in which the creep behaviour was shown to be determined by the intergranular diffusion of cations. The following equation showing

the GdO concentration x (w/o) dependency of (U, Gd)O pellet creep rate e was 3 2

suggeste2 l Mira[2.43]y da b t e i .

log e = log e UO2 - 0.2 x.

It was suggested [2.45] that the effect could probably be due to the higher cation

diffusion rate lesn si s stabl , Gd)0e(U solid solutions. 2

. 2 Cree , Gd)0p (U rat f eo pellet sstresse w withilo e nth s regim < 50-6 e( 0 MPa)s i

almos2 t proportional to stress [2.43-2.45].

wits creeA e hth 2 pelletsp . ratU0 cree 3 e f eo th , p rat f Gdeo 203 dope 2 pelletdU0 s i inversely proportiona secone th o t l d powegraie th f n o rsiz e [2.43, 2.44].

Dat irradiation ao n cree powed pcreean w functioa la r s pa f Gdno 2O3 content have not been published, and hence the standard U0 equations should be used in the absence of

any other information. 2

~~A 6~~

~~10" 30MPa a 8 *10~~

~~10 0 7. 5 6. 0 6. 5.5 10000/T (K'1)~~

~~Fig. 2.17. Temperature dependence of (U, Gd)O2 sample creep rate [2.45] 10* ————————————————————————~~

~~Gd2O3 content (wt %) o o a MP 2 1~~

~~6.0 6.5 10000/T (K'1)~~

~~Fig. 2.18. Temperature dependence of (U, Gd)O pellet creep rate [2.43] 2~~

~~42 TABLE 2.9. THERMAL CROSS SECTIONS NEUTRONF O ABSORPTION, WESTCOTT FACTORS RESONANCEAND INTEGRALS GADOLINIUMOF [2.46]~~

~~Gadolinium Content of Thermal Westcott Resonance isotopes isotopen i cross factor integral natural mixture section (%) (barn) (bam)~~

Gd ( Nat. mixture ) - 488904 ±10 0.8467 390±10 152 Gd 0.2 7350 ±2 0.9784 2020± 160 154 Gd 2.1 85±12 0.9967 230± 26 Gd155 14.8 60900± 500 0.8425 1447±100 Gd156 20.6 1.5±1.2 1.0006 104± 15 Gd157 15.7 254005 1 0±8 0.8510 700± 20 Gd15e 24.8 2.2± 0.2 1.0009 73±7 Gd160 21.8 0.772 ±0. 0.9997 7.2±1

~~TABLE 2.10. EXPERIMENTAL VALUES OF SPECTRAL INDICES IN Gd FUEL WITH LATTICE PITCH~~

~~15.00 ± 0.03 mm, TRIANGULAR ARRAY AND Gd203 CONTENT 3.0 w/o [2.47].~~

235 Type of lattice Lu176 Pu239 u Pu239 Eu151 In115 Mn55 S and absorber - — J— — S"""*"^ ~™ S°*""™~™™™~ S S S~™~~~""" S~ ~~ """"* in central cell Dy164 or 07" ™2W Dy164 Dy164 Dy164 Regular * 1.597 1.650 1.113 1.483 1.143 5.568 1.363 ±0.026 ±0.027 ±0.017 ±0.023 ±0.018 ±0.086 ±0.022

~~( U, Gd ) O2 ** 1.811 2.733 1.368 1.998 1.934 16.67 2.227 ±0.033 ±0.046 ±0.023 ±0.033 ±6.032 ±6.268 ±0.037 Gd/ U3.0= % O/U = 2.00 235 U6.50= %~~

~~* rod is in a central cell absorbe2 O centra) a d n i G s , ri l U cel ( l * *~~

~~43 2.4. NEUTRONIC PROPERTIES~~

~~2.4.1. Cross sections~~

~~Natural gadoliniu followine th s mha g isotopes: Gd152, Gd154, Gd155, Gd156, Gd157, Gd158~~

~~5 15 157 and Gd. 160 Two of these, Gd and Gd, are characterized by large cross sections of neutron absorptio thermae th n i l energy region presence th o ,t whic e f resonanceeo du hs i s (Table 2.9).~~

~~The energy dependence of the total neutron absorption of natural gadolinium decreases much more rapidly than is suggested by the law 1/v within the neutron energy range 0.08- . 0.1eV 8~~

~~2.4.2. Spectral effects~~

~~The spectral indices are functions of many parameters such as lattice pitch, lattice type and gadolinium concentration (Table 2.10).~~

~~2.4.3. Self shielding effects~~

155 157 Estimates of the Gd and Gd resonance self shielding influence on kinf have been performe codS d usineTV (RRe , Russiagth CKl n Federation) [2.48]. Calculations have shown that the effect of Gd155 and Gd157 resonance self shielding is practically negligible at all

~~0.0010~~

~~0.0005 -~~

~~0.0000~~

«o

~~-0.0005 -~~

~~-0.0010 -~~

~~-0.0015 5 1 0 1 5 20 Burnup (MWd/kgU)~~

~~Fig. 2.19. Influence of 6dî5f Gd1sr resonance self shielding Gdd 15San resonance~~

~~integral variations supercell a kon inffor case [2.48]. D without taking into account of self shielding~~

~~44 TABLE 2.11. BASIC CHARACTERISTICS OF Gd ISOTOPE GENERATION BY U^THERMAL NEUTRON FISSION OTHERAND FISSION PRODUCT DECAY FISSION(TOTALOF SUM YIELDS 2.0) [2.48]~~

~~Isotope Yields of Gd from Total Half- Half- Ufe Life Gd Fission Decaf yo Yield Process Fiss. Prod. ofFP ofGd~~

155 3.10 10"4 3.10 10"* 4.96 a stable 156 * 9.010 0 * 9.010 0 15.20 d stable 157 * 2.710 6 * 2.710 6 15.15h stable 158 2.86 10* 2.00 10* 2.00 10* 46.00m stable 159 1.00 10'7 1.52 10* * 1.610 2 18.70 m 18.56h 160 1.52 10'7 9.40 10'7 * 1.010 9 42.00s stable

burnups (Fig. 2.19). This figure also shows the influence of Gd155 resonance integral variation 155 on k,nf. It is seen that a 3% deviation of the Gd infinite dilution resonance integral from the nominal value, which characterize uncertainte sth f nucleayo rresonanc e datth n ai e energy region, does not markedly influence the results of the calculations.

The neutron absorption cross section is not the only characteristic affecting the Gd depletion rateparticle Th fue.alsd e G ear lo e size importantlarga th n o si t e e capturDu . e cross section, the Gd burnup in the fuel pellets proceeds from the periphery to the pellet interior, i.e. at the beginning of the fuel cycle, Gd155 and Gd157 burning in the peripheral layer, shields neutron penetration deeper intfuee oth l [2.49].

Taking into accoun e isotopith t c compositio e presencth d f nan isotopeo e s with resonance neutron absorption cross sections, the development and use of spectral codes for calculation of the burnup of odd isotopes and their generation from even isotopes is important. These problems were discussed at length at the conference on "In-Core Fuel Management" Pinehurstn i , USA, 1986 [2.50]. Firstly face th , t tha capture th t neutroa f nucleue o th nn i s of either Gd or Gd transforms these nuclei into the nuclei of Gd or Gd respectively, 154 156 155 157 means that all other gadolinium isotopes (at least the chain from Gd154 to Gd157) need to be considered in time dependent problem calculations. Besides this, some gadolinium isotopes can originate directly fro fissioe mth n proces forme e decae somb d th n s an f othey deyo b ca r fission products basie Th c. characteristics (yields, half-lives isotopd G f o )e generatioy nb these processes (for U235 fissio thermay nb l neutrons) summarisee ar , Tabldn i e 2.11 (based ENDF/B-Ioe nth V library).

Comparin date gth a describin contributione gth gadoliniuo st m isotope concentration directly from fissio througd nan decae hth f somo y e other fission product followine sth g conclusion reachede sar :

~~(i) Gd158 and successive heavier isotopes can originate directly from fission of U235 t witbu h relativel yieldsw ylo .~~

~~(ii) Gd155 and heavier isotopes can be formed by the decay of other fission products.~~

~~5 accumulatee Th (iii) d yiel rathes i Gdf do 15 r hig relation h (i othe e th nro t gadolinium~~

~~isotopes). However, as Gd155 originates from the decay of Eu155 with a half-life~~

~~5 of nearly 5 years, this contribution to Gd concentratio15 n probably need not be considere reactodn i r calculations.~~

~~(iv) In spite of the facts that, for the other gadolinium isotopes, the contributions from the decay of other fission products are greater than the direct yields from~~

~~fission correspondine ,th g half-live quite sar accumulatee th shor d an t d yields~~

7 are comparable, only normall s i Gd 15 y considere formee b resula o dt s da f o t fissione th process. Thi explaines si d throug difference hth f theieo r thermal cross sections; that of Gd157 is several orders of magnitude higher than that of the other isotopes.

It is worth noting that all the Gd isotopes are stable except for Gd159, and that the yield of Gd157 give Tabln i e fissio2.1o t e 1ndu product Ue 23th 5f so fissio thermay nb l neutrons is 6.26E-05, and the yield from the fission of U238 is 1.7E-04.

~~The effecdepletiod G f pellete o t th nn i s must als takee ob n into accoun thermale th n ti - mechanical calculation (ses eFR e Sectioth f so n 4.2).~~

~~REFERENCE CHAPTEO ST R2~~

[2.1] GOLDSCHMIDT, V.M., ULRICH, F., BARTH, T., Geochemical Distribution of the Elements CrystaV I : l Structur f Oxide eo Rare th ef so Eart h Metals, Skrifter Norske Videnskaps - Akad., Oslo, I: Mat. Naturv. Kl., 5, 1925, pp.5-25.

~~[2.2] SHAFER, M.W., ROY, R., Rare earth polymorphism and phase equilibria in rare earth oxide-water systems" . CeramAm . J ,. Soc., 42(11) (1959) 563-570.~~

~~[2.3] GUENTERT, O.J., MOZZI, R.L., Monoclinic modificatio f gadolinino a sesquioxide, Acta Cryst., 11(10) (1958) 746.~~

[2.4] ROTH, R.S., SCHNEIDER, S.J., Phase equilibri systemn ai s involvin rare gth e earth oxides. Par . Polymorphis1 t oxidee trivalene th th f mf o so t rare earth ions. J , Res. Ntl. Bur. Std., 64A(4) (1960) 309-316.

~~[2.5] BEALS, R.S., HANDWERK, J.H., Solid solutions in the system urania - rare earth~~

~~oxides: I, UO2 -GdO,.5, J. Am. Ceram. Soc., 48(5) (1965) 271-274.~~

~~[2.6] WARSHAW, I., ROY, R., Polymorphism of the rare earth sesquioxides, J. Phys. Chem., 65(11) (1961) 2048-2051.~~

~~[2.7] HAGLUND, J.A., HUNTER, 0., Elastic properties of polycrystalline monoclinic~~

~~. CeramGdAm . OJ , . Soc., 56(6) (1973) 327-330. 2 3~~

~~[2.8] DOLE, S.L., HUNTER, 0., Elastic properties of some GdO-Hf0 compositions, J. 3 2~~

~~Nucl. Mat., 59(1956) 207-214. 2~~

~~[2.9] Handbook of chemistry physics, 56th Edition, CRC, 1975.~~

~~[2.10] Handboo specian ko l refractory element compositionsd san , Metallurgie, Moscow (1969).~~

~~46 [2.11] CASE SMYTH, E. , , HUNTERJ. , , Microcrackin0. , g behaviou f monoclinio r c~~

~~Gd2O3, J. Nucl. Mat., 102 (1981) 132-142.~~

~~[2.12 , S.M.HO ], RADFORD, K.C., Structural chemistr f soliyo d solution U0e th 2-n si Gd203 systems, Nucl. Techn., 73 (1986) 350-360.~~

~~[2.13] WAD , NOROAT. , TSUKUIK. , , "Behaviou K. Gd- , 2 20UO 3 f fuel"o r , Proc. Int. Conf. BNE Nuclean So r Fuel Performance, 15-19 Oct. 1973, London , 63.1UK , - 63.3.~~

~~[2.14] OHMICHI, T. et al., On the relation between lattice parameter and O/M ratio for uranium dioxide-trivalent rare earth oxide solid solution, J. Nucl. Mat., 102 (1981) 40-46.~~

~~[2.15] BEALS, R.Y., HANDWERK, J.H., WRONA, B.Y., Behaviour of urania - rare earth oxides at high temperature, J. Am. Ceram. Soc., 52(11) (1969) 578-581.~~

~~[2.16] UNE, K., OGUMA, M., Oxygen potentials of (U,Gd)O solid solutions in the~~

~~temperature range 1000-1 500°C, J. Nucl. Mat., 115 2 (1983) 84-90.~~

~~[2.17] UNE, K., OGUMA, M., Oxygen potential of U096Gd004(U02-3 wt%Gd3) solid solution" . NuclJ , . Mat. (19851 13 , ) 88-91.~~

~~[2.18] BLACKBURN, P.E., J. Nucl. Mat., 46 (1973) 244.~~

~~[2.19] The Committee on Analytical Chemistry of Nuclear Fuels and Reactor Materials, Japan Atomic Energy Research Institute, JAERI 4053, 57 (1971).~~

~~[2.20] KIHARA, S. et al., Z. Anal. Chem. 303 (1980) 28.~~

~~[2.21] FUKUSHIMA et al., The effect of gadolinium content on the thermal conductivity~~

~~of near-stoichiometric (U,Gd)02 solid solutions, J. Nucl. Mat., 105 (1982) 201- 210.~~

~~[2.22] CHOTARD, A. et al., "Out-of-Pile physical properties and In-Pile thermal~~

~~conductivity of (U,Gd)02", IWGFPT 26, Vienna (1986) 77-86.~~

~~[2.23] NEWMAN, L.W t al.e . , Burnable Poison r Extendesfo d Burnup Fuel Managemen- t Presen Futured an t , 7-2 57-37o t .~~

~~[2.24] THORNTON, T.A t al..e , "Thermal conductivit f sintereyo d urania-gadolinia"S AN , Winter Meeting, Washingto Nov, nDC . 14-18, 1982, 348-349.~~

[2.25] SKOGEN, R.B. KILLGORE, M.R., "Performance of Exxon nuclear gadolinia-bearing fuel in pressurized water reactors", ANS Topical Meeting LWR Fuel Performance, Orlando, Florida, April 21-24 1985, 5-43 to 5-49.

~~[2.26] MILOVANOV, O.A.,PROSELKOV, V.N AL.T .E , Mechanical, physicad an l thermophysical propertie f U-Gso d fuel. , MoscowPreprinKl C RR t , 1994.~~

~~[2.27] INABA t al.e . , H , Heat capacity measuremen f U,.o t GdO (0.00~~

~~330 to 1500K, J. Nucl. Mat., 149 (1987y ) 341-348.~~

~~[2.28] HIRAI , ThermaM. , l diffusivit f U0yo -Gd 0 pellets . NuclJ , . Mat. (19903 17 , ) 2 3~~

~~247-252. 2~~

~~47 [2.29] UNE, K., Thermal expansion of U02-Gd203 fuel pellets, J. Nucl. Sei. Tech., 23(11) (1986) 1020-1022.~~

~~[2.30] CURTIS, C.E., JOHNSON, J.R., Ceramic properties of samarium oxide and gadolinium oxide: X-ray studies of other rare earth oxides and some compounds,~~

~~J. Am. Ceram. Soc., 40(1) (1957) 15-19. [2.31] GROSSMAN, L.N. et al., (U,Gd)0 2 phase equilibria at high temperatures, Colloques Int., CNRS 205 (1972) 453.~~

~~[2.32] MATSUOK t al.e A , "Out-of-pile testinpropertiee th hign e go th hf so conten t gadolinia bearing fuel", 1990 Fall Meetin Atomice th f go Energy Societ f Japanyo .~~

~~[2.33] JAMANOUCHI, S., TACHIBANA, T., Melting temperature of irradiated UO2 and~~

~~U0-2wt% GdO fuel pellets up to burnup of about 30 GWd/tU, J. Nucl. Sei. and 2 3~~

~~Tech.2 , 25(6) (1988) 528-533.~~

~~[2.34] f UOHANSONo -Gde 0Us fue , pressurizen L ,i l d water reactors, Technical 2 3~~

~~Researc2 h Centre of Finland, Report 630 (1986) p. 39.~~

[2.35] Presented in part at the Spring Meeting of the Atomic Energy Society of Japan, April 1992 wor e donjoins a .Th s kwa ea t ventur utilitie5 f eo s (Kansai, Kyushu, Shikoku, Hokkaido and JAPCO) and NFI.

~~[2.36] KOGAI t al.e . , T In-pil, out-of-pild ean e grain growth behaviou f sintereo r dU0 2 and (U,Gd)O2 pellets, J. Nucl. Sei. Techn., 28(6) (1989) 744-751.~~

~~[2.37] LITTLECHILD, J.E., BUTLER, G.G., LESTER, G.W. "The production of burnable poison oxide fuel", Proc. Int. Conf. BNES on Nuclear Fuel Performance, 15-19 Oct. 1973, London , 65.1-65.4UK , .~~

~~[2.38] LOOSE, A. et al., "Diffusion measurement in U0-Gd0", Proc. IAEA Symp. on 2 3~~

~~Improvement Waten si r Reactor Fuel Technolog Utilizationd yan 2 , 15-19 Sept. 1986, STI/PUB/721 (1987) 578-584.~~

~~[2.39] YUDA t al.e , , "UOR. , 2-Gd2O3" Solid State Reacton", Fall Meetin Japae th f gno Atomic Energy Society, Oct. 1991, L473 73 ,~~

~~[2.40] ARAI t al.e . , ElectricaY , l conductivit f near-stoichiometriyo c (U,Gd)0 solid~~

~~solutions . NuclJ , . Mat. (19870 ,15 ) 233-237. 2~~

~~[2.41] HIRAI, M. et al., "Young's modulus of Gd203 - U02, Nb205 - U02 and Ti02 - U02 pellets". Fall Meeting of Japan Atomic Energy Society, Oct. 1991, F58, p20~~

~~[2.42] Wärmeleitfähigkeir PEEHSZu , M. , t Gd-Zusatsen mi Plastizitäd 2 un t U0 n vo t . J , Nucl. Mat., 106 (1982) 221-230~~

~~[2.43] HIRAI t al.e . , "CreeM , p Propert f U0yo 2-Gd203 pellet", Annual Meetin f Japago n Atomic Energy Society, Apr. 1988, L34, p262~~

~~[2.44] WATARUMI t al.e , , "ThermaK. , l propertie f Gdso 0 dope pelledU0 t (l)", Fall 3 2~~

~~Meetin f Japago n Atomic Energy Society2 , Oct. 1988, G366 p3 ,~~

~~48 [2.45] GODIN, J.G. t al.,e , "Mechanical propertie f Uranium-Gadoliniuso m pellets", International Conference on Atomic Power in Space Podolsk, Russian Federation, 21-23 October 19939 1 . ,p~~

~~12.46] MUGHABHAB, S.F., Neutron Cross Sections, v.1, part 8, New York - London, Academic Press (1984).~~

[2.47] POLJAKOV, A.A t al..e , "Investigatio f neutronino c parameter f uranium-wateso r lattice with Gd absorber" in Modelling and investigation of neutronic processes nuclean i r power reactors, Moscow, Energoatomizdat, 118-121.

[2.48] PSHENIN, V. et al., "Comparison of calculations for WWER-1000 lattice containing BA rods with gadolinium", USSR contribution to the IAEA CRP on Safe Core Management with Burnable Absorbers in WWERs (1990)

~~[2.49] HARTLEY t al.e , , TransK. , . ANS (19834 4 , ) 526-527.~~

~~[2.50] ANS Conference on "In-Core Fuel Management", Pinehurst, North Carolina, USA, 1986.~~

~~49 3. FUEL MANUFACTURING~~

~~3.1. FABRICATION PROCESSES~~

~~Fabricatio done b en mainlndiffereno ca tw y yb t methods:~~

~~(ioxide)U moleculaa d t san a mixin d G f gro level~~

~~(ii) mixing of Gd and U oxides as fine powders.~~

~~formee Th achieves ri sol-gee th coprecipitatio d y db lan n processes [3.1 -3.13] while the latter is achieved by dry blending [3.14-3.21].~~

~~3.1.1. Sol-ge coprecipitatiod an l n~~

~~earle th yn I stage f developmenso methoa t f sinteredo 2 pelledU0 t impregnated with gadolinium nitrate solution and subsequent reduction to oxides was tested. However,~~

~~using this metho difficuls wa dt obtaii o t n high (>0.5%) concentration f Gdso 2 OU0n 3i 2, and to ensure uniform Gd distribution in the pellet volume.~~

About three decades ago the sol-gel method was introduced. In this method uraniu d gadoliniuman m nitrate solutio s i gellen n i ammoniad . This providea s homogeneous mixtur uraniuf eo gadoliniumd man structuree Th . f pelleto s obtaine thiy db s method is a solid solution and the grain size is determined by the regime of high temperature treatment. The flow sheet of fuel production is given in Fig. 3.1.

There are two routes for the sol-gel process. One is internal and the other is external gelation. In both cases gadolinium nitrate, Gd(N03)3, and uranyl nitrate UO2(N03)2, are dissolved in water to achieve the correct gadolinium oxide (Gd203) to uranium dioxide (U02) ratioGde 2Th O. 3 content produce shoulth t exceen i dno t % otherwisd10 w ene phase formatio takn nca e place.

~~Powder preparation~~

For the internal gelation route, hexamethylene tetramine (hexa, CN)H is added 4 12 temperatura t a e below 10°C solutioe .Th forcen s i iner throug n s a ga y t db hnozzla e into 4 hot silicon oil at around 90°C. Hexa decomposes at this temperature and gives off

ammonia (NH3) which gelâtes the purged solution droplets within seconds. The gelled microsphere washee sar d first with carbon tetrachloride removo t the silicod e enth an l noi with ammonia solution.

~~Foexternae rth l gelation route, ureammoniud aan m nitrat addenitrate e eth ar o dt e~~

~~solution mixtur uraniu, eof gadoliniumd man . Urea make complesa x with uranyl ions~~

ammoniu(UOd )an , +2 m nitrate provides stability improvo T . physicae eth l propertief so 2 the fuel, titanium trichloride (TiCI3) can be added to yield about 0.1% titanium dioxide (TiO2) in the sintered fuel.

~~Sintering behaviour~~

~~Sinterin carries gi t undedou "nitrogea r n+ hydrogen" r "argoo , hydrogen+ n " atmosphere at 1600 to 1700°C for 3 hrs. Densification is improved by sintering under a wet atmosphere.~~

~~50 URANYL NITRATE SOLUTIO NITRATd G N+ E SOLUTION~~

~~PREPARATION OF SOLUTION WITH RATED U : Gd RATIO~~

~~AMMONIA SOLUTION PRECIPITATION~~

~~WASHING AND DRYING (OPTIONAL)~~

~~CALCINATION AND REDUCTION~~

~~UO2-Gd2O3 POWDER~~

~~PELLET FABRICATION~~

~~Fig. 3.1. Example flow-sheet of Gd pellet preparation by the sol-gel technique~~

~~CONTROL STEPS FABRICATION OPERATIONS:~~

~~2 POWDEUO R GdjOg POWDER~~

~~BLENDING WITH BINDER AND LUBRICANT~~

~~CARBO CONTENd G d Nan T~~

~~PRECOMPACTION, GRANULATION~~

~~HUMIDITY, FLOWABIUTY FILJJNG WEIGHT~~

~~GREEN DENSITY. GEOMETRY SINTERING DENSITY, CHEMICAL and PHASE COMPOSITION, Gd DISTRIBUTION, GRAIN SIZE GRINDING~~

~~GEOMETRY plAMETER)~~

~~H20 AND H CONTENT. RESINTERING TEST, O/MRATÏO~~

~~3.2.Fig. Example fuelGd preparationflow-sheetand blendingQC dry of by~~

~~51 Resulting product~~

Uranium dioxide, gadolinium oxide and titanium dioxide all form a homogeneous phase upon sintering presence Th . gadoliniuf eo m oxide tend increasso t porosite eth d yan lattice th e parameter presence Th . f titaniueo m dioxide tend decreaso st e bot f themho .

~~e sol-geTh l metho t use no n industria i ds i d l production becaus e largth ef o e amount f liquiso d wastes.~~

~~3.1.2. Dispersio f gadoliniuno m oxid uraniuen i m dioxide matrix blendiny bydr g~~

~~Dry mixing of U02 and Gd203 powders mechanically is a much simpler process and is therefore used industrially in all plants. The flow sheet of fuel production is given in Fig. 3.2.~~

~~Powder preparation~~

~~Uranium dioxide powde prepares i r d from reductio f eitheno r ammonium diuranate or ammonium uranyl carbonate (wey routedr ta routes ,y e.g.Integrateb r o ) Routy dDr e~~

(IDR). The physical morphology of dry route U02 powder is very beneficial to the manufactur f non-segregatineo g blends containing gadolinia powder powdee a Th .s ha r ver pouw ylo r densit f abouyo g/cm7 consist0. td ope3an n a f nso structur f interlockineo g platelet type crystallites which prevent segregatio mixture th f no e with gadolinia.

Blending can be achieved by means of local high shear mixing using an orbital screw blender. Batches up to a few tonnes can be fully homogenised with no segregation on discharge. For laboratory work on the 1 kg scale, the two powders may be physically stirred together and passed through a 100 micron sieve three or four times to ensure homogenisation.

additioe Th f TiOno r aluminiuo 2 m oxide (AI2O smaln 3i ) l quantities improvee sth density, grain size and thermal conductivity of the final product. Stearic acid or zinc stéarat usualls ei y adde lubricanta s da .

Granulatio performens i d firs isostatiy b t c compression followe crushiny db d gan sieving granulatA . e siz f les eproducedo s i s m tha m granulat e n1 Th . lubricatees i d with additional zinc stéarat pelletd ean producee sar y uniaxiadb l compression greee Th .n densit f theoretica o abou s yi % 50 t l density.

~~Sintering behaviour~~

A two-step sintering process oxidizin, n firsa n i t g atmospher thet 110d ea nan 0C ° reducina n i g atmospher t 1700°Cea , bese leadth t o s t interdiffusiooxides U d an . d G f no This however canno achievee b t continuoua n di s type furnac industrian uses a ea n di l fabrication plant achieveo T . same th e result, sinterin s performegi humidifiea n di d hydrogen atmosphere at 1700°C for 3-4 hrs producing a slightly oxidizing atmosphere to enhance grain growth and to achieve good interdiffusion of the Gd and U oxides.

~~Resulting product~~

~~additioe Th f gadoliniuno m oxide retards sinterin impartd gan highesa r porosity.~~

~~The O/U ratio also increases with Gd2O3 content. The sintering conditions, like temperature, duration and nature of atmosphere, all affect the physical and mechanical~~

~~properties of the product. At 4, 8 and 12 w/o Gd203 contents, the O/U ratio is 2.03,~~

~~52 ratiM 2.00s oi O/ e 2.092.06d th , d an 2.00, an , 1.9d ,an 9 respectively densite Th . y achieved is 94.5 - 95% of theoretical density.~~

~~By slightly changing the technological regimes, it is possible to obtain a dispersion~~

type structur whicn ei ricd hG h regions wit fine-graiha n solid solutio f higno h Gd203 content (4.5 - 25 w/o) are dispersed in a matrix of reduced Gd203 content and normal grain size (7-13 //m). Studies of the thermal-physical properties have shown that pellets with such dispersed-type structure possess higher thermal conductivity.

When usin powdeR gID r wit hpora e former pore th , e size distributio binomians i l jt/3 m1 d sizesan .witm // Resinterin1 h increases t hr 1700°g a 4 2 densite r Csth fo y yb 0.6%ratie f opeoTh o . closeo nt d porosit affectes yi larga o dt e extensinterine th y b t g atmosphere.

~~3.1.3. Dispersio f gadoliniuno m oxide microsphere 2 matriU0 a x n si~~

Another method of introducing Gd203 into U02 is to prepare Gd203 microspheres of about 100 //m by the sol-gel technique and disperse them in a UO2 matrix. The fuel thus obtained contains separate Gd203 and U02 zones. This fuel is appealing because the matrix is pure UO2, so the melting point and the thermal conductivity are not significantly changed. Moreover, since the gadolinia microspheres have well defined shapes, neutronic calculation more sar e precise. However almoss i t ,i t impossibl achieveo t homogeneouea s

~~distributio f Gdno 203 microspher 2 matrixU0 a ,n e i whic undesirables hi , and, wite hth additional calculational complications, this metho industria n t usea no n do ds i l scale.~~

~~3.2. QUALITY CONTROL~~

The IAEA has recently published a "Guidebook on QC of MOX and Gd bearing fuels" [3.22], with the contribution of BELGONUCLEAIRE, FRAGEMA, PNC and SIEMENS/KWU. Since it still represents the best material on the subject, excerpts of it will be utilized hereafter.

~~3.2.1. Specifications~~

The main problems in Gd fuel manufacture and assembly fabrication are to prevent mixin f pelletgo mixind sensuran o t f rodgd o san e tha locationd ro t fuee th l sn assembliei s are correct. Specific Quality Assurance measures, including specific inspection methods and techniques, are established to control this. Otherwise the specification is basically the same as for standard U fuel with some Gd-specific extensions which will be outlined below.

~~3.2.1.1. Gadolinium oxide~~

~~Typical specifications impose manufacturee th y db [3.22]e ar r .~~

~~- Los f weighso ignitionn o t : leso s w/ tha 2 n1.~~

~~- Impurities (max ppm related to Gd):~~

~~B: 6 ppm F: 40 ppm Cs: 1000 ppm Fe: 300 ppm Ca: 400 ppm N: 600 ppm Cd: 12 ppm Ni: 200 ppm m pp 0 30 : Si m pp 0 4 CI:~~

~~other rare earths (Dy, Eu, Sm, Tb, Yb): 300 ppm~~

~~53 TABLE 3.1. EXAMPLES OF Gd INFLUENCED REQUIREMENTS TO BE INCORPORATED SPECIFICATIONTHE IN~~

~~Oxygen-to-uranium ratio: Ref.~~

~~(2.00= U either00.00O/ + Gd 1o % 2t 8 2 w/x p Ou o30.01+ ) Gd 52 O3 [3.22]~~

~~or O/(U + Gd) = 2.00 + 0.02 [3.22]~~

~~or O/U = 2.020 + 0.045 for5wloGd2O3 [3.23]~~

~~Density Ref.~~

~~+ 0.15 either p = O • 0.04 % Gd O _ (g/cm3) up to 12 w/o Gd 0 [3.22] - 0,20 9 9 0 2 3 2 3~~

~~where p is the specified density of foe gadolinia-doped I/O 2~~

~~p0 is the theoretical density of pure UO2~~

or * -: ***

~~100 = where h t p 2 UO Gd% 2O3~~

~~3 10.25-10.5= 0 5 (g/mor ) for5w/oGdz03 [3.23]~~

~~54 Tablo3.1. (CONT.)~~

~~Gd Distribution Rot.~~

~~relative area optican s(o l micrograph):~~

~~either high Gd2O3 conten < 15w/t o [3.22]~~

~~pure UO2 < 40 % [3.22, 3.23, 3.24, 3.25]~~

~~3 mixeO 2 UOdGd oxid - 2 e soli d% [3.22solutio4 9 - , 0 3.23n6 , 3.25, 3.26]~~

~~or Gd2O3 - rich particles with d > 10^/m shall contain~~

~~less than 2.5% of the Gd2O3 mass [3.22]~~

~~or Gd2O3 - rich agglomerates 40-100 ^m shall be aree lesth a f so tha % n2 [3.23]~~

~~local concentrations:~~

~~average siz f Gdeo 2O ric3- hm ^ agglomerate0 20 o t 0 1 < s~~

~~( to be defined in the specification ) [3.22, 3.23, 3.25, 3.26]~~

~~maximum siz f Gdeo O rich zon < 15/ure n [3.24] 2 3~~

~~average size of UO2 rich zones < 30 to 1000^m definee specificatiob e o (th t dn i n ) [3.22, 3.23, 3.24, 3.25]~~

~~Gd2O ric3- h zone volum % 5 e1 fractioo t 6 n< [3.24, 3.26]~~

~~55 -Gd 155: 14.9 ± 0.6 w/o~~

~~Gd 157: 15.7 ± 0.6 w/o~~

~~- Particl belom o e// w/ size0 w1 5 9 :~~

~~belom o 9// 9w/ 0 w2~~

~~3.2.1.2. Pellets~~

Over and above the requirements for U0 pellets, some Gd-specific requirements 2 have also to be met and controlled: e.g. 0/U ratio depends on the Gd203 content. The 2 equivalendefinitioU0 e th f nto pelle relate e theoreticate b densit th o t o dt s yha l density

~~and the allowable range of distribution of oxide phases (U0, Gd0 and U-Gd-0 solid 2 3~~

~~solution) must als specifiede ob illustrates ,a Tabldn i e 3.1. 2~~

~~3.2.1.3. Fuel rods~~

~~The additional controls specifie ensuro t e dar e tha mixino tn f pelletgo confusior so n of lengt positiod han f differenno t zonepellee th f sto stack occurs.~~

~~3.2.1.4. Fuel assemblies~~

~~The only difference from U FAs is encountered at the final stage of manufacturing, and quality control when it is necessary to ensure proper positioning of the BAF fuel rods.~~

typeo f Tw verificatioso usualle nar y performed througs i e traceabilite On h.th y system othee th , througs i r permanene hth t markin e FRsth f g.o Since bot vere har y simple and fast, one is performed just after insertion of the fuel rods in the loading magazine and the second after loading the fuel assembly just before securing the nozzles.

~~3.2.2. Control techniques~~

~~The control techniques common to U fuel will not be described here. The following are those specifi fueld G :o ct~~

~~3.2.2.1. Pellets~~

~~distributiod G e Th determinens i ceramography db y with adequate colour etching measured an opticay db l contro imagr o l e analysis.~~

~~The GdO content is analysed by mass spectrometry, gamma spectrometry or X- 3~~

~~ray fluorescence2 .~~

~~3.2.2.2. Fuel rods~~

~~correce Th t positionin pelletd G e sth withif go pellee nth absencte stacth d kf an e o rogue U pellets therein are controlled by:~~

~~- activated gamma radiography of the U pellets~~

~~- high field magnetometry of Gd pellets~~

~~- low field magnetic examination~~

~~56 - Visua imagr (o l e analysis) examinatio pellee th f nto stack differene th f i , t pellet types have specific characteristics (e.g. length)~~

~~combinatioa - abovee th f no .~~

~~3.2.2.3. Fuel assemblies~~

~~The first verification is performed by bar code or other marking identification (preferably automatic) relativ positioe th magazin e eo t th nn i e (usually computer recorded).~~

~~The final verification, after loading of the fuel assembly, is performed either by a template or by camera recording to cross-check the position of the Gd fuel rods in the fuel assembly.~~

~~REFERENCE CHAPTEO ST R3~~

~~[3.1] FERGUSON, D.E., DEAN, O.C. HASS, P.A., CEND-153 (vol. 1) pp. 23-38, 1961.~~

~~[3.2] FERGUSON, D.E., Progress in Nuclear Energy, Series III, Vol. 4, Process. Chemistry, pp. 37-78, 1970.~~

~~[3.3] HAAS, P.A., KITTS, F.G., BEUTLER, H., Chemical Engineering Progress Symposium Series. Nuclear Engineering-Par. , pp Vol t , XVIII80 . 63 . No . 16-27.~~

~~[3.4] FERGUSON, D.E., DEAN, O.G., DOUGLAS, D.A., Third Conferenc Peacefun eo l Uses of Atomic Energy, P/237, USA, pp. 307-313, 1964.~~

~~[3.5] HAAS, P.A., CLINTON, S.D., KLEINSTEUBER, I and EC. Prod. Res. Dev. 5(3), pp. 236-244, 1964.~~

~~[3.6] HAAS, P.A., CLINTON, S.D., KLEINSTEUBER, Can . Eng.J . , Dec. . 348-353,pp , 1966.~~

~~[3.7] ZIMMER, E., NAEFE, P., RINGEL, H., Trans. Am. Nucl. Soc., Vol. 20, pp. 591-593, 1975.~~

~~[3.8] ZIMMER, E., RINGEL, H.D., Nucl. Tech. 45(1979) 287-298.~~

~~[3.9] ZIMMER, E., NAEFE, P., Nucl. Tech., 42(1979) 163-171.~~

~~[3.10] YAMAGISHI, S., YOSHIHISA, T., J. Nucl., Sei., 22(1985) 995-1000.~~

~~[3.11] MATHEWS, R.B.P.E.T . NuclHA ,J . Mater., 92(1980) 207-216.~~

~~[3.1 2] GÜNDÜZ, G., ÖNAL, L, DURMAZUCAR, H.H. J. Nucl. Mater., 1 78(1991 ) 21 2-216~~

~~[3.13] GANGULY , BASAK C. ,. Nucl J , .U. , Mater., 178(1991) 179-183.~~

~~[3.14] LITTLECHILD, J.E., BUTLER, G.G., LESTER, G.W., Proc. Int. Conf Nuclean o . r Fuel Performance, London (BNES, 1973) paper 65.~~

~~57 [3.15] DAVIS, H.H., POTTER, R.A., Mater. Sei. Res. 11(1978) 515.~~

~~[3.16] MANZEL, R., DOERR, W.O., Am. ceram. Soc. Bull. 59(1980) 601.~~

~~[3.17] UNE, K., OGUMA, M., J. Nucl. Mater., 131(1985) 88.~~

~~[3.18 , S.M.HO ] , PADFORT, K.C., Nucl. Technol. 73(1986) 350.~~

~~[3.19] UNE, K., J. Nucl. Mater., 158(1988) 210.~~

~~[3.20] YUDA, R., UNE, K., J. Nucl. Mater., 178(1991) 195.~~

~~[3.21] RIELLA, H.G., DURAZZO , HIRATAM. , , NUGUEIRAM. , , R.A., . NuclJ . Mater., 178(1991) 204.~~

~~[3.22] IAEA-TECDOC-584 Bearind G d gan FuelsX , GuidebooMO , f Internationao C Q n ko l Atomic Energy Agency, February 1991.~~

~~[3.23] GODIN, J.G. t al.e , , "Mechanical propertie f Uranium-Gadoliniuo s m pellets". International Conferenc Atomin eo c Powe Spacen i r , Podolsk, Russian Federation, 21-23 October 1993, p.19~~

~~[3.24] NEWMAN , ThermaL , Physicald an l Propertie f Urania-Gadoliniao s , Babcoc& k Wilcox Utility Power Generation, Rep. RAW-1759 (1984).~~

~~[3.25] ASSMANN , PEEHSH. , , ROEPENACKM. , . NuclJ , .H. , Mater. (19883 15 , ) 115.~~

~~[3.26] HAELDAHL , ERIKSONL , . NuclJ , .S. , Mater. (19883 . 15 , 66 )~~

~~58 . 4 DESIG MODELLIND NAN G CONSIDERATIONS~~

~~4.1. PRINCIPLE GADOLINIF SO A FUEL ROD, ASSEMBL REACTOD YAN R CORE DESIGN~~

~~4.1.1. General considerations~~

The purpos f Sectioeo n 4.1.2 - 4.1.4. describo t s i . factore eth s influencine gth choicfued G l f desigeo seenumbes a n a y nb f differeno r t countries reasone th d r an ,sfo their eventual decisions.

~~The design of a first core or reload with Gd fuel requires consideration of a number of parameters:~~

~~a. Distribution of the gadolinia in the pellet (large or small grain size, annular or solid pellet) . Gadolinib a content (weight percent)~~

c. U-235 enrichment of UO2-Gd2O3 d. Axial grading of gadolinia in the rod assemble th n i s . yNumbee FR d G f o r assemble th n i s . f LocatioyFR d G f no g. Number of FAs in the core . Locatioh core th e n i f fres (loadinno s hFA g pattern)

linkeloadine e th ar h do t , g g Point optimize pattere , e ar , sd b nan d durin reloaga d safety analysis with the constant aim of maximizing core operating margins (DNB, LOCA, etc.). Points a, c, d, f are specific to a particular burnable absorber design.

~~4.1.2. BWR fuel design~~

Several companies in the world supply Gd fuel for BWRs including Siemens/ KWU in Germany, GE and SPC in the USA, ABB in Sweden and Hitachi, Toshiba and NFI in Japan. Design feature fued G l f produceo s n Japadi takee nar n fro e Japanesmth e contribution to the BAF CRP and those of GE and ABB from open literature.

Gd fuel has been used in Japanese BWRs from the early 1970s with the adoption of an advanced 7x7 type fuel. Since that time, the fuel design has been changed several time ordesn i increaso rt e reliabilit adapwelo s t ya s changea lo t reacton si r design and operation methods. Gd has been used continously in spite of the design changes - see Table III.1 (App. III).

~~Sinc earle eth y 1980s Axially Zoned Reactivity Fuel (AZR fuel bees )ha n used [4.. 1]~~

thin I s typ fuef eo l assembly U-23e th , 5 enrichmen Gdd 2an tO 3 conten dividee ar t d axially into two parts (Fig. 4.1). In the upper part, the enrichment is 0.2-0.5% higher and Gd content is lower than in the lower part. The number of Gd-bearing rods is not indicated in Fig. 4.1, but with regard to the data given in Tables III.7 and III.8 (App. Ill) the average number of Gd-bearing rods for one FA for an 8 x 8 design is equal to ~ 5-6 for Hitachi fuel fueR l AZ itsel I fuele abls i fTh NF reduc. o e r t fo axiae 7 aneth ~ dl peaking factoy b r flattening the axial power distribution. Reactor operation and PCIOMR has consequently improved.

~~59 Controd Ro l~~

~~Channex Bo l~~

~~A-E : Enrichment (A>B>OD>E) i a.b : Gd203 Content a (a>b)~~

~~d FueRo l 4 5 6 7Gd bearing Type: fuel rod~~

~~Fig. 4.1. Typical enrichment Gofd 3 content2an 0 distribution BWa n i A R F number( indicated position not rods and Gd are of )~~

Since 1988, zirconium lined Zr-2 cladding tubes have been used and in 1992 a new 8x8-4 fuel design was adopted as a high burnup fuel. Design parameters of the typical latest Japanese BWR fuel design are shown in Table 4.1. The fuel structure is shown in Fig. 111.1 and Fig. III.2 (App. III).

~~Typical Siemens/KWU FA BWR designs are shown in Fig.4.2.~~

~~4.1.3. PWRfuel design~~

~~4.1.3.1. Fuel assembly design~~

~~France~~

Until now, the required maximum Gd203 concentration was 8 w/o Gd for 1 7 x 17 FAs in 900 MWe PWR plants, due to utility requirements (fuel management schemes and cycle lengths lower than 18 months). It will also be 8 w/o for the future 1300 MWe PWR new strategy (with an increased cycle length).

~~Concernin FRsd G locatioe g,th e FRAGEM th f no A propose o positioningstw s (Fig. 4.3):~~

~~- insid Fe Aeth - in the peripheral row of the FA.~~

~~60 TABLE 4.1. LATEST BWR FUEL DESIGN (8x8 HIGH BURNUP)~~

~~Item Sub Item Unit Parameter~~

~~Fuel pellet Diameter cm 1.04 Height cm 1.0~~

~~Density %TD 97~~

~~Material U02, UO2-Gd2O3~~

~~Gd2O3 content w/o 6 or less Cladding Outer diameter cm 1.23 Thickness (total/liner) mm 0.86/0.1 Material Zr lined Zr-2~~

~~Fuel rod Fuel effective length m 3.71 Pellet clad gap mm 0.20 (Plenum volume/fuel volume) ratio - 0.1 Initia pressure H l e bar 5~~

~~Fuel assembly Total length m 4.47 Fuel rod pitch cm 1.63 Rod/rop dga cm 0.40 Outer diamete f wateo r d ro r cm 3.40~~

~~Number of fuel rods 60 Numbe f wateo r r rods 1~~

~~Burnup Region average MWd/t 39 500 FA MWd/t 50000~~

~~Max. LHGR kW/sec 44.0~~

~~Max. fuel UO2 °C 1 590 temperature~~

~~UO2-4.5wt%Gd2O3 °C 1 740 Max. clad Outer surface °C 310 temperature~~

~~The choice depends on the use of the gadolinia rods in the core ( boron concentration decrease, power distribution control durin cycle)e gth .~~

With peripheral positioning, the gadolinia efficiency is reduced at beginning of life. However, ther bettea s ei r distributio f gadolinino a effects ovewhole th r e core (fresh gadolinia fuel assemblie surroundind san g assemblies) powee th o S r . peaking factos i r decreased at end of cycle (Fig. 4.4), the most critical moment for Fxy control in loading patterns with gadolinia.

The peripheral positioning is chosen in Sweden { annual cycles require gadolinia for power distribution control) mid-assemble Th . y locatio chosens i Francnn i elonge( r cycles require gadolini r boroafo n concentration reduction).

~~61 uranium fuel assembly MOX fuel assembly~~

~~approx 0 7 e approx 0 4 e, Pu in U-tails~~

~~appro? 09 x U-tailn i u P appros , ?¥ 0 x~~

~~approe 3 0 1 x U-tailn i u P sappro. ë 9 0 x~~

~~Gd+ approë 2 O3 30 1 x U-(aiIn i u P sappro, e 3 1 x~~

~~enriche, approë Gdo + i xd OU average content of fissile material (f i 2 3 uraniue th Un i m o fuew/ 3l6 assem~~

~~fueX 235 l assemblMO n i blyu P . y equivalent with respect to reactivity)~~

~~Fig. 4.2. Typical distribution of fissile material and Gefe O3 in ATRIUM™ 9 fuel assemblies for symmetric BWR lattices~~

~~235~~

~~The U matrix of a U02-Gd203 pellet is mostly depleted uranium. Power peaking~~

5 is hardly dependent on the Uenrichmen23 t of the matrix and, from an economic standpoint, a depleted uranium matrix is little different from an enriched matrix when the numbe f freso r h gadolinia bearing rod cyclr spe relativels ei y low.

Axial gadolinia concentration grading is not necessary from a hot spot control point f axiallo e yus truncate e allows oth f t vieFR bettesbu d a w d, G r utilizatio f gadolinino a (lower residual penalty).

summaryn I maie th , n feature f FRAGEMA'positioningso d (G A F d sG , enriched matrix determinee ar ) e fuee basith th lf o managemensn o d t characteristics it d an s associated gadolinia use. usualls i R F yd G homogeneou e Th t FRAGEMsbu alss Aoha already designed assemblies with axially truncated Gd FRs.

~~Japan~~

Design studies showed that 6 w/o Gd in U02 was the optimum in terms of reactivity control capability at the begining of cycle, peaking factors and residual penalties. This was confirmed from usage experience. Therefore, only one content of Gd203 (6 w/o) is used in Japanese PWRs.

~~62 MID-ASSEMBLY LOCATION~~

~~PERIPHERAL LOCATION~~

~~Ü Instrumentation thimble H Guide thimble • Gadolinium rod~~

~~Fig. 4.3. Positioning of Gd FRs in a FA~~

Melting temperature and thermal conductivity of U0-GdO both decrease with 2 2 3 increasing Gd2O3 content. The local power peaking factor (Fq) of a Gd FR therefore must be less than 60-70% of the design Fq of a standard FR. The Fq limitation is achieved by limiting the radial power peaking factor of a Gd rod to less than 80% of the assembly averag time th epoisot d ea whe G e nn th disappears limitatioF e Th . determinens i y db

evaluatin meltineffece e gth th f o t g poinq t reductio Gdo nthermae Ow/ ) (70°th , 6 r Cfo l 3 conductivity reduction and th2 e required safety margin. In order to satisfy these requirements . RecentlyFR uraniue d ,th G o ,mw/ enrichmen6 a r fo o reducew/ s ti 5 1. y db out-of-pile experiment resultE PI d s demonstratesan d that safety margin sufficiente sar .

Fig. 4.5 shows vsk . burnup for 17 x 17 fuel. The ok,f Gd fuel is 1 ~ 2% lower than F F N tha f standardo t fue poisold wheIN G e nn th disappear sU w becaus23lo 5 e enrichmenth f eo t and residua poisoningd G l Ue 235Th . enrichmen f standaro t do fuew/ designes 1 i l 4. e b do t

~~for 13 EFPM, 2.5 ~ 3 reload batches, to account for the low kof the Gd fuel assembly. f in~~

desige Th n parametere x 15,1ar threr 5 - 1 sfo 17 , e7 x type14 x fuelf so 4 s-1 shown in Table 4.2. Basically the fuel design depends on the fuel vendor and consequentl ytypeo thertw f desige so ear eacr nfo h fuel typ indicates ea d Tabldn an i 2 e4. in Fig. III.3 (App. III). Fig. 4.6 shows Gd FR arrays commonly used by fuel vendors in Japan.

~~63 Mid-assembly location Peripheral location 1.56 -~~

~~1.46 -~~

~~0 2000 4000 6000 8000 10000 12000 Cycle bumup (MWd/tU) Fig. 4.4. Comparison between peripheral mid-assemblyd an locations: Fine power distribution (Fxy)~~

~~A K-inF f o f 1.40~~

~~1 30 -~~

~~1 20 -~~

~~1.10 -~~

~~1.00 -~~

~~0 90 10000 20000 30000 40000 50000 Fig. 4.5. K-inf vsburnup (17x17, 4.1 w/o)~~

~~64 TABLE 4.2 JAPANESE PWR FUEL DESIGN~~

~~Item Type 4 1 x 4 1 15 x 15 17 x 17~~

~~Fuel pellet Material U02/UO2-Gd2O3 U02/U02-Gd203 U02/U02-Gd203~~

~~235 U enrichment U02 fuel w/o 4.1-3.4 4.0-3.4 4.1-3.6~~

~~U02-Gd203 fuel w/o 2.6-1.9 2.5-1.9 2.6-2.1~~

~~Gd2O3 content w/o 6 6 6 Density %TD 95 95 95 Diameter mm 9.29 or 9.21' 9.29 or 9.21* r 8.058.1o 9 ' Height mm 11.2 or 10.0* r 10.011.o 2 ' 10.0 or 9.0'~~

~~Burnup Region average MWd/U t 41 000-30 000 43 000-30 200 000-30 4 0 300 Maximum FA MWd/U t 48000 48000 0 4800 Maximum pellet MWd/U t 62000 62 000 62 000~~

~~Max. fuel temp. at rated power U fuel °C 2 000 1 870 1 770 Gd fuel °C 1 980 1 840 1 730~~

~~Transient U fuel °C 2 350 2270 2 270~~

~~Gd fuel °C 2 170 2 170 1 990~~

~~Cladding Material Zr-4 Zr-4 Zr-4 Outer diameter mm 10.72 10.72 9.50 Thickness mm 0.62 or 0.66' 0.62 or 0.66* 0.64*r 0.5o 7~~

~~Pellet claddinp gga mm 0.19 0.19 0.17~~

~~Max. surface °C 350 350 350 temperature~~

~~OS Two Designs are in use (See Fig. III.3, App.Ill) I/I TABL 2 (CONT.E4. )~~

~~Item 4 1 x 14 15 x 15 7 1 x 7 1~~

~~Fuel assembly Number of fuel rods 179 204 264 Fuel rod length m 3.86 3.86 3.9~~

~~- Fuel rod pitch mm 14.1 14.3 12.6~~

Length m 4.06 4.06 4.1 Size mm 0 20 x 0 20 214x 214 214x214 Number of grid spacers 7 or 8 7 9 Material of grid spacer Ni-Cr-Fe alloy Ni-Cr-Fe alloy Ni-Cr-Fe alloy Number of RCCA guide thimble tubes 16 20 24

~~Outer diam. of RCCA guide thimble tube mm 13.69 13.87 12.2~~

~~Thicknes f RCCso A guide thimble tube mm 0.43 0.43 0.41~~

~~Material of RCCA guide thimble tube Zr-4 Zr-4 Zr-4 Numbe f instrumentatioo r n tubes 1 1 1~~

~~Outer diam. of instrumentation tube mm 10.72 13.87 12.2~~

~~Materia f instrumentatioo l n tube Zr-4 Zr-4 Zr-4 (14X14 Type) (15X15 Type) (17x17 Type)~~

~~EH------Instrumentation tube El------Control rod guide thimble tube • ••••••• Gd FR Q...... R F u .~~

~~Fig. 4.6. Positions of Gd FRs in a Japanese assembly~~

~~Korea, f Repo ,~~

As the nuclear programme expanded, Korea was interested in "kuksanwha" (indigenizatio f foreigno n technology f variouo ) s aspect f commerciaso l nuclear reactor technology pars thif A .o t s policy, fuel desig manufacturd nan e starte fore th joinmf o dn i t projects with foreign vendors. These joint projects include fueld dG .

In orde keeo t r moderatoe pth r temperature coefficient negativ reduciny eb e gth required soluble boron concentration, gadolinia integrated burnable absorbers have been chosen instead of discrete burnable poison rod assemblies. Some FRs are replaced by Gd FRs with the fuel pellets containing 6-9 w/o gadolinia in enriched uranium.

In a typical reload core design in Korea four or eight Gd FRs are used in a gadolinia poisoned FA. The position of Gd FRs in a poisoned FA and the position and number of poisoned FAs in the core are optimized for better power peaking control and cycle economy [4.2].

Becaus e absorbeth e s integrai re fuel th , o t gadolinil a integrated burnable absorbers offer a lower residual penalty and thus lower fuel cycle costs than the standard burnable poison rod assembly. Also the poisoned fuel assemblies can be loaded under control assemblies for better design flexibility. Moreover, the separate handling and disposal costs associated with standard burnable poison RCC assemblies are not necessary gadolinia n i a integrated burnable absorber design.

~~Sweden~~

The use of advanced features, such as LLLPand axial blankets, is readily achievable with ABB PWR Fuel [4.3]. Both improve the fuel cycle cost but normally reduce the available margin for both LOCA and DNB. This has been overcome by the use of axially graded gadolinia integral burnable absorber (BA) [4.4]. This concept, patented by ABB Atom, makes it possible to use both advanced features simultaneously without having to increase safety limits.

~~67 Core • —\~~

~~Gd-rod Fuel assembly Thimble tube Fuelrod~~

~~fuel rod Every second pellet is a Gd-pellet~~

~~No Gd-pellets~~

~~A>ially graded burnable absorber,BA~~

~~Core Height Top~~

~~blanketNo BA no Blanket no BA Blanket full length gadohma Blanket axia/ly graded BA~~

~~Bottom Relative power~~

~~The influence axialon power distribution witn axial i graded BA~~

~~Fig. design 4.7B PWa AB .f o R fuel assembly~~

~~68 ThdesigA eB n (Fig. 4.7) control powee sth r distributio thren i e ways:~~

~~t reduceI • powee sth r capabilit frese th hf yo fuel .~~

~~distributet I • powee sth r more evenly amongs assembliee tth ove cycld e srth an e length.~~

~~• It distributes the power more evenly along the length of the assembly.~~

concepA B e lone basefavourabls Th d i t th gan n do e experience gaine BWRsdn i , but adapted to the specific needs and requirements of PWRs. This means that, for example rodA uraniue B onls s i th , e yth slightlf mo y enriche dbees (1.3ha n o used.8w/ ) This concept has already been implemented in three consecutive PWR reload designs and has resulted in substantial fuel cycle cost savings.

~~4.1.3.2. Core design~~

With the continous aim of maximizing core operating margins (DNB, LOCA, etc.) FRAGEM developes Aha wida d e rang f fueo e l managemene tMW scheme 0 90 r fo s reactors (12 feet cores) and 1300 MWe reactors (14 feet cores) for EOF plants and foreign plants, with the following principal characteristics:

~~frese enrichmenth A hF • t (U23 varie) 5 s from 3.25 4.00%%o t ;~~

~~• the composition of the reload varies from 32 assemblies per reload in Ringhals (900 MWe assemblie4 6 o )t reloar spe thirn di d core fuel managemen lonr fo t g cycles (1300 MWe).~~

~~Since 1983, about 5200 Gd FRs have been tested under severe operating conditions (load follow, reduced power operation) and have demonstrated improved PWR operating flexibility.~~

~~Japan~~

~~Requirements for safety certification of a reload core are specified by the regulatory guide, which is enacted by the Japanese Atomic Energy Safety Commission. The issues to be considered are:~~

• Stuck rod margin • Maximum LHGR • Maximum FA burnup • Maximum reactivity insertion rate • F N rXY • Moderator temperature coefficient • Doppler coefficient (not require routinr dfo e reload core) • Contro dro d wortd ro p ro lr incidenhfo t

~~• FAH f° contro drod ro pl incident~~

~~r • ControN l rod worth for rod ejection accident~~

~~• FQ for rod ejection accident.~~

69 Typical reload core designs of 4, 2, and 3 loop PWRs for 13 EFPM equilibrium cycles are shown in Appendix III. In most cases about 50% of the FAs are Gd FAs. Additionally smala , l numbe f conventionao r l usee BPRAb supplemendo y t sma fued G l a t cor reduco et e power peaking factor. core Mosth ef o tdesign OUT-Ie sar N refuelling schemes, but some cases are IN-OUT or hybrid.

~~Korea, Rep, of~~

~~i. Reload core~~

As an example of a reload core design, Kori-2 cycle 7 is shown in Appendix IV. The core consists of 121 FAs. In refuelling (low leakage loading pattern), 52 standard FAs are currently being replace fres4 y db h standard fuel assemblies U-enrichment(3.4o w/ 1 d )an fres8 4 h KOFAs (Korean Fuel Assemblies designe KAERy db KWU)d an I .

~~Out of these 48 KOFAs, 28 fuel assemblies contain U0 fuel rods of 3.5 w/o U- 2 enrichmen fue0 2 l d assembliean t s contaif no U0 o 2w/ -Gd 6 - 20 3o fuew/ 8 l rod1. f so~~

~~Gd0 The length of cycle 7 is 14,42 GWd/tU (375 EFPD). This compares with the 3~~

~~cycl2 e 6 length of 13.1 GWd/tU (348 EFPD).~~

~~. Initiaii l cores~~

initiae Th l cor f Yonggwang-3/eo currentls 4i y under design jointl KAERy yb d an I characterizes i C-E t I . batc4 a leakag w y hlo db e fuel management scheme, enrichment f part-lengto e us e zoninth h d controgan l rods althoug t finalized ye desige t hth no n I ns i . the current proposed design, the Gd FAs contain Gd FRs of natural uranium and 4 w/o

Gd203. The initial cycle length will be 12 months. The design goal for the average burnup 5 GWd/ti5 s r lateUfo r cycles. Fig.4.8. show e infinitth s e multiplication factors a s functions of burnup for the 8 shimmed fuel assemblies. The factors for 12 shimmed FAs are shown in Fig. IV. 17 (App. IV).

~~1.0 10 20 30 Burnup (GWd/tU)~~

~~Fig. 4.8. Infinite multiplication factor for YGN-3/4 initial core: 8 Shim FA~~

~~70 1000~~

~~6 8 10 12 14 Burnup(GWd/tU)~~

~~Fig. 4.9. YGN -3/4 critical boron concentration: w/4 o GdiOs natural- uranium~~

~~Burnup~~

~~p To Bottom Top Bottom~~

~~Gd shimmed core Boron shimmed core~~

~~Fig. 4.10. Axial power distribution vs burnup~~

71 Th fractiond eG s remainin functions ga f burnu9 so 1 showd e pan ar 8 Fign 1 i . .IV (App. IV). Fig show9 .4. criticae sth l boron concentratio functioa s na f burnupno . Fig. 4.10 illustrates the evolution of the axial power shape of the Gd-shimmed core as burnup progresses in comparison with the axial power shape of the conventional boron-shimmed coreworts i t I .h noting thaaxiae th t l power peak appear- Gd e scycle th latth f eo n ei shimmed core [4.5].

~~The following observations and conclusions are noted:~~

• Occurrence of high peaking factor in mid-cycle due to too rapid depletion of Gd (low Gd content). f hig I contend hG useds i t• residuaa , l reactivity penalt occury yma . This would shorte cycle nth e length.

~~• The benefit in fuel cycle cost appears to be minimal, if not negligible.~~

~~• More sophisticated analysis requires i optimizeo dt :~~

~~number of Gd FRs in an assembly assembln a n i s locatioFR y d G f no position of Gd FAs in the core Gd content enrichmen FRs d f Uo G t23 n 5.i~~

~~4.1.4. WWER fuel design~~

~~4.1.4.1. Fuel assembly design~~

~~Gd FR design is similar to U FR design in terms of pellet size, rod diameter, etc. Uranium enrichment values in Gd FRs are traditionally lower than in a U FR and are defined~~

~~by thermo-physical characteristics. Gd2O3 content is determined by a number of requirements for the core including fuel cycle length, power peaking factors, reactivity coefficients, etc.~~

For the three-year WWER-1000 fuel cycle, average enrichment is equal to 4.23% (66 fuel rods in the peripheral rows with corner FRs of the next row having an enrichment of 3.6% and all other FRs having an enrichment of 4.4%). In the Gd FRs enrichment was

~~reduced to 3.6% [4.6]. The range of Gd2O3 content was chosen to be from 4 to 9 w/o and the number of Gd FRs per FA from 1 2 to 24. All cases were calculated during 1991-~~

~~93. A FA design with 18 Gd FRs (8 w/o Gd203) was selected for the first loading of Gd fuel into WWERs (Fig. 4.11).~~

~~In choosing the position of the Gd FRs in the FA, allowance is made for the following effects:~~

~~- the non-uniformity of the FR power distribution and the position of the highest- power FR;~~

~~reducee mutuae th d th an d - s l depletio"shadowingFR d G n i e nd th ratG f f o e"o them.~~

~~- the control and protection system efficiency.~~

~~72 0~~

~~o) - guide tube~~

~~(V) - central tube~~

~~Fig. 4.11. Fuel arrayd ro f WWER-1000 o A F~~

A basi bees c ha loadin n A chosenF a g n i patter s ; witFR nd h G witthi e 8 hs1 th highest rated FR is at the FA periphery up to a burnup of ~ 16 MWd/kgU and then moves bees ha n^ reducewheA F, F e centre ne th th 1.0o th d t f o t eo .

~~4.1.4.2. Core design~~

Calculations have shown the acceptability of the basic design with regard to power shaping in the reactor core. The use of a refuelling strategy with a LLLP is complicated for WWER-type reactor difficultieo t e sdu s with smoothin f radiago l peaking, i.eo .t

~~constrain the FA power peaking factor within FQ < 1.35.~~

The first experience was obtained at Kola NPP Unit 3 with a WWER-440 reactor [4.7] when 36 FAs with an enrichment of 3.6% were moved after the third year to the fourte corth f ho e yead periphery f irradiationen o r e th y burnuB .e ,th p reache GWd/t8 d3 U (average burnup for three years fuel cycle around 28 GWd/tU). During this period the relative power of the highest rated FA was maintained within the limits (FQ< 1.26). Further reductio f neutrono n leakage using suc reloaha d scheme coul achievee db fuly db l loadin core th ef g o peripher y witwhics hFA h have already been burne r severadfo l years

~~innee th e corern i th par f .o t This would complicat smoothine t wouleth bu y x dF f go provid savinea fresgn i h fuel requirement alsd reductioosan a f radiationo n damage th o et reactor vessel.~~

~~73 3 16 I 2 16 I 1 16 I 0 16 I 9 15 I 8 15~~

~~7 15 I 6 15 I 5 15 4 15 I 3 15 I 2 15 1 15 0 15 I 9 14~~

~~139 I 140 I 141 142 143 144 I 145 146 147 I 148~~

~~103 I 104 I 105 I 106 I 107 I 108 I 109 I 110 I 111 I 112 I 113 I 114 115~~

~~16 17 I 18 I 19~~

~~1 I 2 I 3 I 4 I 5 I 6~~

~~Gd FA~~

~~Fig. 4.12. FA array f WWER-1000o core~~

The economic benefit for the transition to a LLLP for a WWER-440 is around 2-3% of the fuel cycle cost for equilibrium fuel cycles, although it is more appreciable for transition fuel cycles (up to 10%).

Design studies for the transition to LLLPs for WWER-1000 plants have also been performed somn I . e plants, partial LLLPs have already been used whereby some burnd tan som e placecore eth ar efres n s o dperiphery hFA . Wor underwas i k validato yt e eth introduction of more complete LLLPs.

In 1993, 12 WWER-1000 FAs with Gd fuel were manufactured. Each FA includes wit s fueld 1 showh8s G FR a , Fign i . 4.11. Thes wers eFA e loade WWER-100e th dn i 0 of the third unit of Balakovo NPP, as shown in Fig. 4.12.

~~The fuel enrichment in the Gd FRs was 3.6% and the Gd content was 8 w/o. s wa s Enrichmen 4.4%s enrichmenFR e f thesd wa Th o otheG s e .s th e e FA FR rth n i t n i t~~

~~set to prevent overheating at the end of the reactor cycle (Table 4.3); kinf vs burnup for standard FAs and FAs with Gd FRs are shown in Fig. 4.13.~~

Analysis of the calculations shows that substitution of lumped boron burnable absorber by Gd fuel allows a WWER-1000 core pattern with reduced radial neutron leakage, thus enabling an increase in fuel burnup and a decrease in fast fluence to the pressure vessel compared wit standare hth d three-year e fueth f l o campaigne us e Th . integral burnable absorber also improves reactor stability (self-regulating properties) due increasn a o t negativn ei e feedbac coolana kvi t temperature resulting fro mdecreasa n ei boroe th n content.

~~74 TABLE 4.3. TYPICAL WWER FUEL DESIGN~~

~~Standard WWER-1000 Characteristics WWER-1000 FA with FA Gd fuel~~

~~Core average LHGR (kW/m) 16.7 16.7~~

~~Number of FRs in FA:~~

~~- U FRs 312 294~~

~~- Gd FRs 18~~

~~Number of guide tubes 18 18~~

~~Fuel rod: - length (cm) 353 353 - outer diameter (cm) 0.915 0.915~~

~~- cladding material b N % 1 Z - r b N % 1 Z - r~~

~~- maximum LHGR in U FRs (W/cm) 448 448 - maximum LHGR in Gd FRs (W/cm) 260 - enrichmen s (w/FR oU U-235 f o t ) 4.4 4.4 - enrichment of Gd FRs (w/o U-235) 3.6~~

~~-Gd203 content (w/o) 8 - maximum fuel temperature in U FR (°C) -1400 -1400 - maximum fuel temperature in Gd FR (°C) -1360~~

~~Fuel pellet: - outer diameter (cm) 0.753-0.757 0.753-0.757 - central hole diameter (cm) 0.24 0.24~~

~~4.2. MODELLING~~

~~4.2.1. Thermal and mechanical considerations~~

Gd conten consideres i t havo dt e four significant effect fuen so l performancet i : degrades fuel thermal conductivity t producei , distortedsa , rapidly changing radial power profile t i reduce, fuee sth l melting t i slowpoin d fissioe an stth releas s nga e ratee Th . codes develope countrien i d s producing and/or usin fued gG l take these effects into account.

~~75 lumpe8 1 + 4. d4% absorber rods ( B content 0.036 g / cm3 )~~

~~4.4 % + 18 Gd FRs ( 8% Gd2 O, )~~

~~1.20~~

~~1.10 -~~

~~1.00 10 20 30 Burnup (MWd/kg)~~

~~Fig. 4.13. Kinf versus bumup WWER-1000a n i~~

~~Belgium~~

e COMETHTh E d fueperformancro l e modelling code developey b d BELGONUCLEAIRE contains models for which several options can be selected to fit the specific characteristic modellede fueb e th o t luseeve f e n so modeTh ca n rn. inpuow s l hi t for the fuel rod constituent (e.g. cladding or pellet) which departs from the models contained in the user manual.

For most common fuels, BELGONUCLEAIRE has incorporated adequate models in the basic library and the user manual includes preferred options recommended for each of e commerciath l fuels used routinel n NPPsi y . These recommendation e basear s n o d benchmarking against a large experimental data base.

For gadolinia fuel, the data base includes, as of December 1992, a total of 31 FRs, made by different fabrication processes and irradiated over a large range of operating conditions thin O s. basis leadino tw , g phenomen fuee th lf abehaviouo r were identified and incorporate e COMETHth n di E modellin fuel d e dependencG th :f o g f thermao e l conductivity upon Gd content and the evolution of radial power profile.

singlA e conductivity correlation, independen f whetheo t mads fuee th rwa y l eb coprecipitation or mechanical blending, has been found to model the fuel adequately, homogeneite th largelo t e ydu y achieve industriae th y db l fuel manufacturers [4.8].

acceptabln A e mode evolutiof o l n with burnu radiae th f po l power profile alss soha been achieved. Early sensitivity studies [4.8 indeed ]ha d demonstrate majoe th de rthib so t factor influencin modelline gth g result f gadoliniso a fuel.

~~76 Experience has also shown the importance of adequately modelling the axial evolution of burnup, which depends on good predictions by the nuclear design code of the axial depletio f gadoliniumno .~~

Four fuel vendor Japan si n have theicodn ow re package analyso st thermae eth l mechanicad an l behaviou fueld G .f o rDetail f modelso f individuaso l code packages vary, principln i t bu followine eth addressede gar :

~~pelletsr fo - : densification, swelling, temperature distribution, fission producs ga t release, thermal expansion, creep~~

~~r claddingfo - : creep (therma irradiation)d an l , oxidation (inne outer)d an r , hydrogen pick-up, stress, strain, thermal expansion, crud deposit~~

~~- for rods: PCI, gap conductance.~~

The code package analyses these inter-related items step by step. The code packages have been continually revised based on the PIE results to date. Recently, the rim effec bees ha tn attracting attention, especiall fued G l r pelletsyfo .

Three code r WWER-100performancR sfo F d G d an e0U calculations, namely PIN- modl, the system RETR and START-3 have been developed respectively by RRC "Kurchatov Institute" and ARSR Institute of Inorganic Materials. In comparison with earlier codes, they have now taken into account the effects of extended burnup, Gd impact on thermal-mechanical propertie radiad san l depressio f neutrono n functio fluxa s a f burnuno p (calculated with neutron-physics codes).

In the PIN-mod1 code, the increase of gas release from the fuel at extended burnup is taken into account (the gas release model is changed in the region of grain growth during fuel restructuring recoile share th ;th o f FGt e o mechanis e Rdu mincreaseds i ; fast neutron flux from adjacen alss i s o accounteFR t d for)

The PIN-mod1, RETR and START-3 codes were tested on the results of post- irradiation examinatio f WWER-100no [4.9]s 0FR .influenc d AlsoG e th ,n therma eo l expansion coefficient, on density of fuel and on power density distribution over the fuel pellet radiu takes swa n into account.

The code adopted in the UK for FR performance calculations by BNFL and Nuclear Electric (NE, formerly CEGB) is ENIGMA [4.10], developed jointly by BNFL and CEGB. The cod s constructei e a modula n di r fashion, enablin modelw ne g s describing specific mechanisms, suc thermas ha l conductivit releases incorporatega e b r yo o t , tested dan d easily, without upsettin code'e gth s overall calculational robustness. Modificatioe th f no code to allow modelling of non-standard fuel types, such as gadolinia-doped, niobia-doped or mixed oxide fuels, has consequently proved relatively straightforward. ENIGMA has therefore been extende includo dt suitablea e thermal conductivity formulation specifi,a c radial power profile model for gadolinia-doped fuel, a modified melting point equation and a modified fissio diffusios nga n model.

~~77 For standard U fuel, the expression used to calculate thermal conductivity at absolute temperature T, is~~

~~k = (a + b*T)~1 + electronic contribution~~

expressioe Th valis i r temperaturek dfo r nfo r temperatureFo s. K abov 3 e77 s takes i constan a K applys no i t 3 " 77 "b ,t t valua whice k th f , eo beloK h depend3 w77 s on the host material and "a" depends on impurities, "a" is therefore proportional to burnup

~~(representing the build-up of fission products in the U02 lattice), such that~~

~~- BurnupC + 1 )( « 0 a = a~~

The coefficient "C" was fitted to thermocouple data from FRs irradiated to intermediate burnups in the Halden reactor. The same trend was subsequently found to fit high burnup temperature measurements from re-instrumented rods at Riso.

~~For Gd fuel, "a" is further modified such that the thermal conductivity equation becomes~~

~~k = [c- (1 -f- d- G) + e* T]"1 + electronic contribution~~

~~weighe wherth s i etG percentag constante f gadoliniaagaies ar o i e c d nd san an ,d a burnup-dependent term~~

~~• Burnup , C + )1 ( - 0 c = c~~

~~expressiono tw equal e e expressioe ar th k Th , . r Fos=0 fo G r valins i gadolinir dfo a weight fractions up to 12 w/o.~~

~~Radial power profiles are calculated using a modified version of the RADAR routine,~~

~~[4.11 ] widely used in fuel performance codes to calculate radial rating profiles for UO2 fuel. RADAR is a physically based code that calculates the radial variation of thermal neutron~~

~~fueflua n xi l pellet irradiatioe Th . n dependenc radiae th f elo power profiles arises through~~

239 5 buildu e depletioe th th d f Pu,p o an bot f U nwhicf o h23 o modellee har RADARy db e Th . modified RADAR models the additional radial rating depression due to gadolinia and the burnuprincipae th f po l gadolinium isotopes with irradiation. Since gadolini vera s ai y strong neutron absorber, radial power profile strongle sar y affected.

~~With these additions modifiee ,th d versio f ENIGMno A (know ENIGMA-Bs na s ha ) been use modeo dt l gadolinia data fro BELGONUCLEAIRe mth E organized international GAIN project, with good agreement.~~

~~France~~

~~FRAGEMA, FRAMATOMhavA eCE completee th d E an extensiv n da e programme of development tesA t CE facilities, basee th n do , whose main objective determineo t e sar :~~

~~- Gd fuel physical-chemical characteristics (melting point, thermal conductivity, etc);~~

~~fued G l thermal-mechanica - l characteristics.~~

~~78 4.2.2. Neutronic Considerations~~

~~Belgium~~

BELGONUCLEAIRE utilizes an improved version of LWR-WIMS, modified on the basis of critical experiments conducted in the VENUS facility (on a national and, with the GAP International Programme, international basis) and from lessons learned as a result of Gd fuel utilization in LWRs, including cores containing MOX fuel.

Four fuel vendor Japan si n have theicodn ow re packag analyso et nucleae eth r propertie fuel d additionn I G .f o s , some utility customers, including their subsidiary companies, have theicodn row e packages. Analytical model individuaf so l code packages are different from each other. In principle, the models are simplified transport theory and ENDF B-VI/V nuclear data are used. However, based on the results of critical experiments and experience of commercial reactors, the packages have been continually revised to improv accuracye eth .

The following is an example of the upgraded methodology of a nuclear code package "HAMMER-AIM" [4. : 1Mor2] e detailed calculational mesh division fued G l a n si pellet compared wit pelleU necessarhe a ar t y becaus vere th yf eo larg e neutron capture cross section of Gd. For this reason, the standard nuclear design code could not be used for Gd fuel. The HAMMER-AIM code which is the multi-group space dependent neutron transport evaluation code in the FR cell has now been modified. Various critical experiments containing Gd FRs were evaluated using the modified HAMMER-AIM code which was also used to assess Gd fuel irradiation data in overseas PWRs. Both evaluations sho validity e analysiwe th th f o s method. This modified HAMMER-AIM code was also applied to the evaluation work of the OHI 2 Gd fuel demonstration irradiation, which verifie validitye dth f thio s analysis metho operatine th r dfo g plant.

During startup physics tests at BOC 5, the measured parameters such as critical boron concentration, MTC, control rod worth and power distribution agreed with the predicted values very well. Also the measured critical boron concentration vs. burnup agreed with the predicted curve. Fig. 4.14 shows the mean power change of a Gd FA vs. burnup, from which it can be seen that the measurements and predictions agree very well.

~~Assembly power~~

~~.o''S'~~

~~, B-''8' •-——— Prediction (Bank D-215 step) 02CY5. GdAssy, C-03. N-13 ) (A ) CO °'th~~

~~t i i i t_t__ i t i i r _ _i 5000 10000 15000 Cycle Burnup (Mwdrt)~~

~~Fig power. A 4.14F cycles d v G . burnup~~

79 r thermafo s mechanicad A an l l modelling FRAGEMA, FRAMATOMA CE e th d Ean have completed an extensive programme of development, based on CEA's test facilities. The neutronic methods for cross-sections and fine mesh calculations have been validated. The qualification is based on two types of experiments:

~~- critical experiments (zero burnup) - depletion experiments~~

~~Korea, f Repo ,~~

e computeTh r codes thausee o perforar t n Kore di E y KAERC e b amd th an I neutronic design of the Yonggwang Unit 3 and 4 reactor intial cores containing gadolinia are DIT and ROCS/MC. Fig. IV.20 to IV.23 (App. IV) show the code system for FA and core design of Yonggwang Unit 3 and 4.

The DIT code is a transport theory based code which performs spectrum and spatial calculation fuesn i l fueceld lan l assembly geometries. Thes calculationT eDI s providw efe group neutron cross section PDQ- the ROCS/Msfor and 7 C codes groufew , p spatial calculations in exact fuel assembly calculations, and isotopic depletion calculations for every a fuecel n li l assembly including subrogion f eaco s h cell when necessary. Calculations performed wit e ROChth S code provide reactor power distributiond an s effective neutron multiplication cod factorsmodulC a ROCM e s i e th en STh i . code which provide two-dimensionaa s l fine mesh local powed fluan x r distribution calculation extracted from the global ROCS results. The DIT and ROCS/MC codes and their methodologies have been reviewe approved dan USNRCe th y db . burnabla s a d G e f absorbeo e neutronie th us n e i r Th c desig reactoa f no r core causes some difficulties in the calculations because of the very large neutron absorption cross sections of the Gd155 and Gd157 isotopes. These very large neutron absorption cross sections cause significant spatial self-shieldin spectrad gan l variatio e thermath f no l neutron flux within a Gd FR. This spatial self-shielding varies strongly with burnup. Modification ROCS/Md an s T havDI C ee codebeeth o t n accommodato s t E mad C y eb e these neutronic effects in order to maintain accuracy. A discussion of these modifications follows.

e mathematicath n i s i changee Onth T f DI eo l o smodellingt . This changs ewa necessar maintaio yt n accurac grou1 calculation4 n yi T p DI FAsneutro d e G r Th . sfo n cross section library was found to calculate reaction rates and infinite multiplication factors accurately over a normal depletion range for a gadolinium reactivity holdown larger than design value y aboub s o (wit factoa exceptionte tw f hon o r ) when compared with calculations wit reference hth grou5 e8 p neutron cross section library exceptioe Th . ns i a non-negligible differenc rhodiue th n ei m reaction rates correctioA . n facto bees ha rn uses i r constructind dfo an derive E C y db g CECOR coefficient librarie r reactosfo r cores containin FAsd determinegG E C . d tha fla0 t1 t frofluo t mx8 region necessare sar o yt model Gd depletion in a FR. CE also determined that additional thermal groups are necessary to perform the few group DIT fuel assembly calculations with the total number f groupo se depletio increaseth r Fo n . d7 calculation froo t m4 predictor-correctosa r method is employed to keep the number of time steps manageable over a depletion range. This predictor-corrector method was checked against calculations with very small time step f acceptabl showo d e san b no t e accuracy neutrod G e nTh .cros s sections generated

~~by DIT for the coarse mesh calculations are functions of the initial gadolinia content, the~~

~~5 157 concentratio soluble f eitheth no d e an rGdr boroo Gd 15 n concentratio coolante th n i . Differentia crosd G l s section coefficients with respec changeo t moderatosn i r densitd yan~~

80 fuel temperatur e alsar eo constructe a functio s a df eitheo ne Gd th rr 15Gdo 5 157 concentration eithed moderato,an e rth r densit square th r yo e fuee rooth lf temperaturo t e depending on the differential cross section. The table-sets that are constructed result in point reactivity errors, due to linear interpolation and the assumed functional dependencies, of les reactivitn i s tha% 1 n0. y wit hrandoa m distribution two-dimensionae Th . l fine mesh MC module of ROCS employs a simpler functional dependency in the Gd table-sets that is use calculato dt e local power burnupd san s withi froA resulte F m nth a s obtained from the global ROCS calculations.

ROCe Th S code include snumbea f provisiono r accommodato st e burnableth e absorber gadolinia t explicitlI . y isotoped traceG e sth s Gd154, Gd155, Gd Gdd 15an 6157 . Other Gd isotopes are treated as a lumped residual. ROCS depletion of Gd has been change maintaido t n accurac commonle th r yfo y used burnup step interva f 100o l 0 MWd/t b computationaya l strategy that permits time step average cross section evaluatee b o st d empiricallan sa y weighted averag f theieo r f eacvaluebeginnino e d th h t sen a tim d egan step. This Gd depletion procedure allows CE to calculate Gd reaction rates within 1 % of reaction rates obtained with calculations performed with very small burnup intervalso N . changes wer emodulC requireM e f ROCeo th do t S since thimodulC sM e obtains neutron cross section datcalculated aan d results directly fro ROCe mth S code.

The few group structures for the DIT fuel assembly calculations were changed from determineFAsgroupE 7 d C G o t .r 4 sfo d tha minimua t fla8 f mt o flu x regions were required for modelling a Gd-U fuel pellet. Some simplifications were made to the cross section representations of the Gd isotopes. For all of the approximations that were made reactoe th o t r design calculations performe ROCS/MCd d an wit T h determineDI E C , d that accuracy of computed quantities with respect to more detailed calculations was maintained USNRe Th .reviewe s Cha ROCS/Mchangee d dth an T sDI Cmade codeth eo t s and their methodologies to accommodate the use of Gd in PWR cores. These changes are typical of those type made by the industry for designing gadolinia cores. The numerical results that were provided show that acceptable agreement has been obtained between detailed calculations and design calculations.

~~codee Th s use Koredn i KAERy ab Siemens/KWd an I nucleaUn i r reload fuel designs including using Gd, are described in App.IV.~~

~~Neutron physical properties of the Gd core are calculated using a neutronic code package [4.13] with the following main features:~~

~~- three dimensional core power distribution , fuel burnup, reactivity coefficients and effectiveness of reactivity regulation are calculated using the BIPR-7 program;~~

~~- a library of small group constants is compiled using the KASSETA-TVEL and KASSETA programs;~~

~~- the power distribution for each rod is calculated using the PERMAK program~~

BNFL utilize e standarth s d Westinghouse Nuclear Design code package ALPHA/PHOENIX-P/ANC which includes the capability to model Gd fuel and which has been validated against operating plant data. BNFL has further validated this code package against data obtained fro VENUe mth S facilit Belgiumn yi .

~~81 REFERENCES TO CHAPTER 4~~

~~[4.1] ENOMOTO, T., "Present status and future perspectives of BWR core-fuel design" (Journal of the Atomic Energy Society of Japan, Vol. 26. No. 2, 1982)~~

~~[4.2] SOHN, D.S., KIM, S.H., YANG, C.K., "Korean nuclear fuels in the nineties", Proceeding KAIF/KNh 6t e th f Sso Annual Conference, 529-539, April 1991, Seoul, Korea.~~

~~[4.3] LINDELÖW t al.e . , "Asea-AtoU , m Advance ReloaR dPW d Fuel Assembly Design", Nuclear Europe, May 1985.~~

~~[4.4] LINDBERG, M. et al., "Advanced Axial Gradation of the Asea-Atom PAAD Fuel", Proceedings of the ANS Topical Meeting on LWR Fuel Performance, April 17-20 (1988), Williamsburg, Virginia, USA.~~

[4.5] SONG, J.S., "Analysis of effects of Gd203 content and UO2 enrichment in Gd burnable poison rod nuclean so r desig f fueno l assembly", RT-NI-YG8-89006K, Korea Advanced Energy Research Institute, July 1989.

[4.6] ONUFRIEV, V.D., PROSELKOV, V.N., SIMONOV, V.D., "Issues of ensuring safety and increasing the economic indices of water cooled reactors in the USSR in the us f U-Geo d fuel", presente RCM firsF e th BA tt d,a 22-25 October 1990, Vienna, Austria.

~~[4.7] PROSELKOV, V.N., et al., "Some aspects of fuel cycle efficiency increase for WWER-440", Preprint IAE, No. 5252/4, 1980.~~

~~[4.8] BILLAUX, M., BLANPAIN, P., BOULANGER, D., "Evaluation of the thermomechanical properties of high burnup gadolinia fuel", ANS Top. Mtg. on LWR Fuel Performance, Williamsburg, April 1988.~~

~~[4.9] BIBILASHVILI, Yu.K., ONUFRIEV, V.D., PROSELKOV, V.N. et al, "Thermo-physical characteristics of WWER-1000 fuel rod in Unit 5 at Novozonezh NPP", Atomnaja Energia, 1993, Vol , Issu. 74 , 450e5 .~~

~~[4.10] KILGOUR, W.J., "Capabilitieal t e , validatiod san ENIGMe th f no A fuel performance code" ANS Topical Meeting, Avignon, April 21-24, 1991, 919-928.~~

~~[4.11] PALMER, I.D., et al, "A model for predicting the radial power profile in a fuel pin", Proceeding f IAEso A Specialists Meeting, Preston, March 1982.~~

[4.12] MATSUOKA, Y., et al, "Gadolinia fuel development and plutonium recycle experience in Japanese PWR", Transactions of the Intern. Symp. on Achievement of Good performance in Nuclear Projects, April 17-20, 1989, Tokyo, 77-79.

~~[4.13] NOVIKOV, A.N., et al., "Code package for WWER cores analysis and some aspects of fuel cycle improvements", VANT, 1992, issue 1, 3-9.~~

~~82 5. EXPERIENCE WITH GADOLINIA FUEL~~

~~5.1. UTILIZATIO GADOLINIF NO A FUEL~~

This section provides a description of Gd fuel utilization worldwide including representative experiences in test reactors and in power reactors.Table 5.1 shows an overvie statufuee d th G l f f wutilizatioo so n whic describehs i d hereafter.

~~5.1.1. Belgium~~

Between 1973 and 1987, the BR3, an 11 MWe PWR was extensively used to test on a reduced scale FAs containing new or advanced fuel types. In the framework of the Belgia programmeD nR& , relatively large quantitie fued G lf witso h increasin contentd gG s

~~TABLE 5.1. OVERVIEW OF Gd FUEL UTILIZATION AND/OR DELIVERY~~

~~Country Status Numbef ro Numbef ro Reactor Starting (fuel vendor) date Uses GdFR d GdFAs type year section~~

~~Belgium Oct93 354 70 BR3 (prototype PWR) 1974 5.1.1 3144 365 PWR 1982~~

~~Germany Feb2 r9 69524 12572 BWR 1974 (KWU) 12954 1396 PWR 1982 5.1.5~~

~~France Dec 1993 5200 600 PWR 1983 (FRAGEMA) 5.1.2~~

~~Korea Oct93 3728 640 PWR 1990 5.1.3~~

~~Japan 1 9 d mi 30 - Test reactor 1984 5.1.4 7092 460 PWR 1984 59980 10228 BWR 1976~~

~~Russia Oct93 34 - MIR (test) 1991 5.1.8 216 12 WWER-1000 1993~~

~~Sweden Sep3 t9 90000 na BWR 1975 (ABB) April 94 5100 na PWR 1984 5.1.6~~

~~UK Dec 93 na 400 BWR 1975 (BNFL) 5.1.7~~

83 were loadee reactoth n di r (Fig. V.1 o acquirt ,) AppV .databasea e thermalth n o e - mechanica nucleae l behaviouth n o fuerd d characteristicG an l f ro cora f seo heavily loaded fued G l [5.d an 1 5.4]- wit X referencmaie U hd .MO Th n an characteristic s e FR d G e th f so FRs are summarized in Table V.1 (App. V). The peak burnup ranged from 33 to 70 GWd/te peath kd W/cmtypica0 LHGUA an 45 .R o t l froevolutio0 m10 e pea th kf no reactoo powetw r fo rr cycle functioa s irradiatioe sa th f no n tim presentes ei Fign di 2 V. . (App. V). The PIE performed on experimental fuel at the request of the licensing authoritie reveales sha speciao dn l features.

Taking advantag existine th f eo g facilitie Belgiae th t sa n Nuclear Centre (SCK/CEN) Moln i , parallel international programme swerP suc eGA GAIs ha conducted Nan d dan various types of Gd FRs were irradiated [5.3].

Gd has also been utilized at the Tihange power plant in all three units as shown in police TablTh thif y. o (App2 progressivel so V) e g V. plan . o t s i t longeo yt r reactor cycles. The Gd FRs represent 0.9-1.9 % of the total FRs loaded as compared with the 7% in the BR3 demonstration cycles. The reloads have been delivered by various fuel vendors (FRAGEMA, KWU etc.C SP ,)

~~5.1.2. France~~

FRAGEMA, FRAMATOME and CEA have completed an extensive programme of development base tesn do t facilitie mai e t CEAsa Th n. objectives have bee investigatno t e physical-chemical characteristics, thermal-mechanical characteristics, neutronic characteristics and qualification of fabrication processes. The neutronic methodology has been qualifieo typetw f experimentso n s o d : critical experiments (zero burnupd an ) depletion experiments largA . e amoun f measuremeno t t dat availabls ai confiro et e mth suitabilit desige th f yo n codes [5. 55.6]- .

~~Since 1983, FRAGEMA's Gd fuel with mid-assembly or peripheral [5.7] positioning schemes has been incorporated in:~~

~~reload3 s wit hdemonstratio8 n assemblies~~

~~30 reloads with from 8 to 36 Gd FAs~~

~~firs1 t cor FAsed witG . 4 h2~~

The fuel management strategies cover a range of U enrichments up to 3.8%, and a variety of fuel management schemes. A typical hybrid type fuel loading scheme (transition to LLLP) in Ringhals 4 is shown in Fig. 11.1 (App. II). Future Ringhals reloads will use Gd FAs in 100% IN-OUT loading schemes.

By the end of 1990, about 5200 FRs had been irradiated, a large proportion of which were under severe operating conditions (load follow, reduced power operation), demonstrating the flexibility of reactor operation with Gd fuel.

~~5.1.3. Korea, Republif co~~

In Korea, 9 nuclear reactors are currently in commercial operation: 8 PWRs (2 of 600 MWe and 6 of 900 MWe) and a CANDU (600 MWe). In addition, 2 PWRs of 900 MWe eac undee har r construction. Until recentl vendoe yth eacr fo r h reactor suppliee dth fuel. Now, KAERI (Korea Atomic Energy Research Institute) is responsible for the fuel and

~~84 TABL£ 5.2. Gd FUEL UTILIZATION STATUS IN KOREA~~

~~Fuel type Plant name r pe s FR d G f o . No w/o Gd and % U235 Reload cycle (year) from which coren i s )(NoFA f .o FA/total Gd FRs in core in Gd fuel first Gd FAs were introduced~~

~~14 x 14 Kori - 1 4/64 6.0/1.80 11 ('90) (121)~~

~~16 x 16 Kori - 2 4/80 6.0/1.80 7 ('90) (121)~~

~~17 x 17 Kori -3/4 —. | 9.0/1.80 5 ('90) Yonggwan2 1/ g - 4 and 8/112 4 ('90) Ulchin-1/2 — 2 ('90) (157)~~

~~(All above o th F e plant suppliee sar ' y db i/Vh)~~

16 x 16 Yonggwang - 3/4 8/14d 4an 0 4.0 0.7/ 1 1 ('94) ( Supplied by KHKAECE ) oo Ui reactor core design and KNFC (Korea Nuclear Fuel Company) is responsible for the fuel fabrication activities. Three type f Koreaso n Fuel Assembly (KOFA) were developed jointly with r Siemens/KWUfo r Kor7 fo 1 i4 Unix 1 7 x r Kort 1 fo 1,14 d i1 6 : Uni1 an 6 2 tx the other six PWRs [5.8].

The KOFA has several significant differences from the Wh and FRAMATOME fuel designs which were used previously in Korean PWRs. The KOFA design includes a thicker cladding, removable top and bottom nozzles, Zircaloy intermediate spacer grids with Inconel corne hold rdan springs s dowFR nd G ,spring s wit additionan ha l leaf.

The KOFAs were loaded firstly into Kori Unit 2 in February 1990 and subsequently othee th int l roal PWRs. Irradiated KOFAs were first inspecte Yonggwann di g Unin i 2 t March 1991. Similar inspections will follow in other units. For the verification of KOFA performance, selected FAs from each type of KOFA were pre-characterized during their fabricatio wil examined e nb lan detaidn i l during refueling shutdown period relevane th f so t units.

~~Table 5.2 shows the Gd fuel utilization status in Korea.~~

KAERI has contracts with KWU for reload core fuel design for all operating reactors and with ABB/CE for initial core fuel design for the Yonggwang 3 and 4 units. More details of these and the experience with other reactors are presented in App.IV.

~~5.1.4. Japan~~

Gd fuel has been used in Japanese BWRs from the mid-1970s with the adoption advancen oa f 7 typd7x e fuel. Since that tim fuee eth l desig changes nha d several times in order to increase burnups and safety margins. The number of Gd FRs in a FA, Gd content, uranium enrichmen assemblarran R a F n yi d an t y hav l beeeal n changed ovee th r years. The design modifications are summarized in Table III. 1 (App. III). Recently an typJ B e 8 fuebees x ha ln 8 introduced. Nowadays, mos f Japan'o t s s FA reloaR dBW contai fuelR FRs,d BW nleaG r dFo . test assembly irradiatio havE PI e d beenan n carried resulte outTh . s showe fueld G integrit.e R dth Detail BW givef e yo s ar section i n 5.3.1 and in [5.91.

Utilization of Gd in PWRs started in the middle of the 1980s. For reload fuel, about half of the assemblies contain Gd. Prior to the commercial use of Gd FRs in PWRs, irradiation experiment werE PI e d carriesomn san o t ed ou experimenta l reactors. Lead test FAs were then irradiated in a commercial reactor.

By 1991, more than 7000 Gd FRs had been irradiated in PWRs and more than 60 000 Gd rods in BWRs without failure. Details of the irradiation experience of all the givee ar Appn s i FA .d IllG , togetheR BW Japanesd ran witR he PW typica l core loading detail r 4-loopsfo , 2-loo 3-loo d plantsR pan pPW .

~~5.1.5. Germany~~

havd containinTess G eo FR tbee w/ 4 n g2- irradiate Obrigheie th n di m reactop u r to maximum burnups of 18 GWd/tU and 33 GWd/tU) [5.10]. LHGRs ranged from 140-180 W/cm to 200-250 W/cm respectively. Gd FRs showed identical dimensional behaviour with regard to cladding creepdown and FR growth as U FRs and fission gas release was low «4%) whic alss n agreemeni ohwa t with comparabl FRsU e . Metallographie examinations revealed the fact that the as-manufactured pellets contained areas of high

86 gadolinia content and some areas of almost pure U02. At burnups above around 25 GWd/tU, large pores appeare o develo dt e area th f previouslt o sa p y higher gadolinia concentration, accompanie somy db e grain growt informatioo h(n numbee th o t r o s r na frequency of these pores is given). These pores were not prone to densification; the Gd fuel was therefore observed to densify less than the U fuel. However, this pore formation stoppe dGWd/tU0 betwee3 d henced an 5 ,an n 2 , becaus relative th f eo e stabilit f thesyo e pores with increasing burnup, it can be assumed that the doped and undoped rods exhibit the same swelling behaviour above around 30 GWd/tU.

A large quantity of Gd fuel (Table 5.1) has been loaded in BWRs since 1970 [5.11] with up to 6.5 w/o Gd; a peak burnup of 52 MWd/kg U has been achieved. Furthermore, Gd fuel with 3 to 7 w/o Gd (and test FRs up to 12 w/o Gd) have been loaded in 7 PWRs achievo t e low-leakage core patterns pea;a k burnu MWd/k5 4 beef s po ha ng U achieved .

~~5.1.6. Finland and Sweden~~

A Finnish-Swedish high burnup fuel evaluation programme [5.12] was initiated in 1984 to investigate a number of phenomena, included in which were a number of Gd FRs burnupd an d G s MWd/kgU8 o rangin4 w/ o t 9 highese p gu o witt Th releass .p hu ga t e seeobserves s 12%A nwa . radiay db axiad an l l gamma-scannin mobile th f go e fission nuclide Cs137, Gd pellets do in fact retain mobile fission products better than pure U pellets even though the Gd pellet is apparently operating at a higher temperature. As releases fro centrae mth l region pellete th knowf e equaso e s ar b volatilr no fo lt e fission products such as caesium and for fission gases, fission gas release should also be reduced. Since it is concluded from this study that less fission product migration takes place, this should counteract the increase in gas release expected as a result of higher fuel temperatures.

The globa9 4 ld experiencGd-fuean R d AtoB G BW o n lm AB extendo f w/ e o 0 1 o st MWd/k averagR F gU e burnup (Fig . manufacturine 5.1)Th . burnud gan p experiencf eo ABB Atom PWR BA fuel is given in Fig. 5.2. The maximum BA assembly burnup is 45 MWd/k. gU

~~5.1.7. United Kingdom~~

~~BNFL's interesburnabld G n i t e absorber fuel originated around 1970 when laboratory scale work was carried out over a range of concentrations up to 30 w/o gadolinia.~~

Production-scale work commence n 197di 2 when BNFL e madth fued r G efo l 0 kg/yea25 o t r Dodewaaryeap 6 oveu 1 t ra ra periodR dBW . Initially e gadolinith , a afte levew/o rros 2 w/o 1981979 but In le1 to Gd 4 . additiothe som w/o at e2.7 nwas s madfue lated wa l an esmala rduple d G l o irradiation xquantita w/ fue r 4 fo lf o y n experiment was also fabricated. Since 1975, over 2800 kg of BWR Gd fuel has been irradiate oven 0 di FAs 40 r, eac f whicho reaches hha s desigdit n burnup peae Th k. assemblies have achieved burnup excesn s i GWd/t0 3 f so U wit failureso hn .

~~5.1.8. Russia~~

A batcexperimenta4 3 f ho (enrichmens FR U-235o d G w/ l contend 6 d ,3. tG an 4 t4. 5.0 w/o, rod length 250 mm) was fabricated and commenced irradiation in the MIR test reactor in 1991. The maximum burnup reached was 4.6 MWd/kgU (FR average = 2.3 MWd/kgU) and the maximum LHGR was 84 W/cm.

~~87 40000~~

~~30000 -~~

~~JS 20000 - z 10000 -~~

~~6 3 0 4 —— I—i pj I | | l I 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 Gd content (w/o)~~

~~20000~~

~~16000 - •sS 12000 - o kfco_~~

~~8000 -~~

~~4000 -~~

~~44 10 i i r i i i i i ii r i i 0-4 4-8 8-12 12-16 16-20 20-24 24-28 28-32 32-36 36-40 40-44 44-48 48-52 Rod Averag) U g k e/ d BumuMW ( p~~

~~Data valid 1993-09-15~~

~~Fig. 5.1. BWR Gd Experience of ABB Atom~~

~~88 Numbe rodd G sf ro Numbe fabricatef ro d rods CO S S 8i § o o c C O O c ) C l 888 1,1, 1,1,1,~~

~~^ ~ i" 9 -~~

~~? - 8 • —~~

~~3 °° tn 2 - to JO M~~

~~no »> 9" o>~~

~~8 I !i- CO [DO: m o O > K m 1 . •Q (D ° •5. s " " ^ ^ . •.-.-.- %^ > < !S ,g 10 ^s- " IK -—.to 9 ^ ê 3 c0» e~ lÄ~~

~~e_ S- in o to~~

oo The first loading of 1 2 demonstration Gd FAs into Unit 3 Balakovo NPP (WWER- 1000 durins autumwa e ) gth n 199 f technicao 3t reloadse e l documentTh . transfer sfo r of this unit to full Gd fuel cycle is in preparation.

~~5.1.9. USA~~

~~Exxon, lateC r SP ANF w ,no~~

Over 530 wits h0FR gadolinia concentrations rangin havo w/ e 0 gbee1 fro o nt m1 irradiate burnupo t GWd/tp 0 du 5 f so U wit failureo hn r problemso s [5.13, 5.14]. From examinations performed on some 8 w/o Gd FRs in Tihange 1 after burnups of 13.2 and 37 GWd/tU differenceo n , clan i s d integrity were noticeable fued G l e showeTh . d less densification, possibly due to the lower operating temperatures on the first cycle of operation. No redistribution of gadolinia within the pellets was observed and the pellet cracking patterns were simila standaro rt pelletsdU wers ,a e change pellee th n si t density.

~~BWFC~~

BWFC have irradiated Gd FAs in the Oconee reactor to a burnup exceeding 58 GWd/tU. PIE was carried out on some FAs and the results confirmed the integrity of the Gd FRs at high burnup. BWFC also supplied Gd FAs for the TMI-1 reactor for the cycle due to start operation in 1993.

~~5.2. NUCLEAR DAT NEUTRONIA- C MEASUREMENTS~~

~~qualificatioe Th f neutronino c code sinterpretatioe relieth n so f threno e typef so experiment:~~

~~Critical experiment tesn i s t reactors that allow definitio f locano l power distributio. n insidFA e eth~~

~~Startup tests and periodic measurements performed in power reactors (boron concentrations, control rod efficiency, 3D power distributions).~~

~~t celHo l examinations (isotopic concentration f adjaceno fued d G an l f sU o t fuel, micro gamma-scanning).~~

~~Information from these three type f experimentso describes si d below.~~

~~5.2.1. Test reactor experiments~~

Such experiments have been carried out in Japan [5.15, 5.16], in France, where FRAMATOM havA CE e d completeEan da progra f mqualificatioo n base critican o d l experiments (zero power depletiond )an s [5.171 Belgiumn ,i , where BELGONUCLEAIRd Ean SCK/CEN have conducted experiments both in the critical facility VENUS at Mol and in the BR3 reactor (e.g depletioe .th n experiment [5.3]P sGA , already mentioned Russin i d aan ) [5.18].

~~90 5.2.2. Neutronic experience from power reactors~~

~~5.2.2.1. France~~

Core physics tests performed during startups and measured data obtained throughout operating cycles have yielde largda e numbe f dat o rsuc n ao h parameters sa FA powers poweD 3 , r distributions clusted ,ro r control worth borod san n concentrations.

For example, Table II.2. (App. II) presents a comparison between measurements and calculations of the initial start-up tests of RINGHALS 4 cycle 11. It also shows cycle length comparison . Fig. 11 d II.r cyclesfo an 2 compare) II (App0 s1 . powed sG r trendt sa t fulho l powe RINGHALr fo r analysie Scycl4 Th . ethesf 9 s o e datcomparisond aan s with predicted design values demonstrates the adequacy of the design models and codes developed by FRAMATOME and used by FRAGEMA for Gd cores.

~~Operation surveillanc f demonstratioeo n assemblie proves sha accurace dth f o y power distribution calculations (agreement power)A F sd betteG .n o r o thaw/ n2~~

~~5.2.2.2. Japan~~

~~Demonstratios nFA~~

~~Demonstration irradiatio physicd nan s test result reportee ar s 2 fro i [5.16dn i m Oh , 5.19, 5.20].~~

Eight demonstration Gd FAs were loaded into the Ohi 2 reactor cycle 5, in July orden i 198 confir4o t r irradiatioe mth n performance .s containeA EacF FR d d hG G 6 d1 Uo 23w/ 5 6 enrichment1. (sed e an Chapted G , o whil) wit4 w/ r he6 enrichmene th f o t standard U fuel was 3.2 w/o. The cycle 5 started in July 1984 and ended in September 1985, reachin cyclga e burnu f aboupo t 15000 MWd/tU.

During the startup physics tests at BOC, the measured parameters such as critical boron concentration, moderator temperature coefficient, control rod worth and power distribution were found to agree very well with the predicted values. Also the measured critical boron concentration and assembly power vs. burnup as shown in Fig. 111.16 (App. Figd . an Ill111.1) 7 (App. Ill) respectively agreed well with predictions.

~~Commercial operation~~

Commercial use of 6 w/o Gd FAs in PWR plants was started in 1988. Eight FAs were loade Mihamn di loo2 ( pa1 core with 14x14 fuel assemblies assemblie2 3 d an ) s were loaded into both Ohi 1 and 2 (4 loop core with 17x17 fuel assemblies). The physics parameters measured durin startue gth p physics test durind san g commercial operation agreed very well wit predictionse hth .

Tables III.7-8 (App. Ill) summariz startue eth p test experienc commerciae th f eo e us l of Gd FAs. They show that there are no nuclear design problems in the large scale commercial use of gadolinia fuel.

~~5.2.3. Hot cell neutronic examinations~~

~~Determination of the isotopic concentration of irradiated Gd fuel has been performed severan i l countries, e.g. Belgium [5.3], Franc Japad e an [5.7n] [5.21].~~

91 Hot cell examinations have been performed on Gd fuel irradiated in Ohi 2, including axial gamma-scanning, Fig. 111.18 (App. Ill), and micro-gamma scanning of pellet cross- sections. Micro gamma scanning revealed the effect of a water cell adjacent to a Gd FR. The water cell promoteresulta s e A burnu fissio th ,d . depletio e U dan th pf d no G f no profile was asymmetric with respect to the centre axis of the pellet, as shown in Fig. 111.19 (App. III).

~~5.3. THERMAL-MECHANICAL BEHAVIOUR~~

~~5.3.1. Overview~~

~~Knowledge of the in-pile performance of Gd fuel results from its steadily increasing use poweove R pase yearn th r PW te tr powen plantsi R extensivd BW s an r n i plant e eus s for 30 years.~~

Several sources of data are available from the open literature and a short summary of the irradiation conditions, reactor type, gadolinia content and gas release is provided in Table 5.3 for PWR conditions. Test fuel rods containing 3-10 w/o gadolinia have been, beinge oar r , irradiate condition R GWd/t0 7 PW o t n di p U su peak burnup with LHGRs irradiateR F d G failur o steadn n di a f w eo rangin yno kW/m5 o t 4 g p o t froU .6 m1 conditions has been reported in the open literature with nominal LHGRs and no specific proble bees mha n revealed eithe MTRn i rnuclean i r so r power plants principae Th . l post irradiation test results of fuel rods referenced in Table 5.3 are reported below and the main features discussed.

fueR l irradiatio BW result e e Th th f so n test presentee sar [5.9]n di , basee th n do result Japanesf so e investigations characterizee Th . wers d8 designFA e typica8x R . BW l They were designe Toshiby b d a Corporatio e firs th s par tna f o reloat e d th fue r fo l Fukushima Daiichi No. 3 Reactor, an 800 MW(e) BWR. The FA consisted of sixty-three FRs and one water rod. Their design specification is as listed in Table 5.4. The ten characterized FAs were irradiated from October 1977 for a maximum of four cycles. A total of seven characterized FAs (two each discharged after 1 cycle, 2 cycles and 4 cycles and one assembly discharged after 3 cycles), were shipped to the hot laboratory of Nippon Nuclear Fuel Development Co., Ltd. (NFD) and examined. The exposure of the assemblies ranged from 6 GWd/tU to 29 GWd/tU which covers almost the entire range over which Japanes fueR l eBW assemblie currentle sar y irradiated.

~~5.3.2. Dimensional changes of the FRs~~

~~5.3.2.1.PWR FRs~~

No significant differences in fuel length changes have been observed between Gd and U FRs irradiated at high [5.2, 5.4] or low [5.9, 5.10] LHGR. The same quasi-linear relationship of rod growth with the fast neutron dose accumulated by both FR types has also been observed [5.4, 5.8, 5.9, 5.15].

e BelgiaInth n experiment t higa s h LHGR claddine th , g creep-dowe th o t e ndu coolant pressure was less pronounced in the Gd FRs [5.2]. For Gd and U FRs, the creepdown phenomenon generally stop arount sa GWd/t8 d4 U correspondin appearance th go t e of ridging whic [5.2 s morhs i FR , ed 5.4] markeG e . th n di

~~Japanese investigations, carrie firs e t durinth d tou e reloadingth Mihama-e th f go 3 NPP, showed a continuous cladding tube creep down [5.9].~~

92 Demonstration irradiation on Gd FRs extracted from one lead test fuel assembly reacto2 i Oh loader e revealeth n di slightls d wa thaoutes e yth t FR rd diameteG e th f o r smaller tharodU nse thath [5.15] f o t . This fac explaines i t difference th e y db th n ei creep dow oxidd nan e film thickness causepowee th y db r difference betweeo tw e nth types of FRs (20 kW/m and 16 kW/m for Gd and U FRs respectively).

In conclusion, at low LHGR (<_ 20kW/m) there is no significant difference between Gd and U fuel. More pronounced fuel-clad mechanical interaction and less creep-down is observed in high LHGR fuel rods, due to the increase of the centre temperature induced lowerine by th thermae th f go l conductivity.

~~5.3.2.2. BWR FRs~~

Every characterized FR (Table 5.4) irradiated in the FUKUSHIMA reactor [5.9] was measure r lengtd fo oute d han r diamete profilometryy b r . Fig.5.3 shows tha growte tth h of measure almoss wa s proportion i t dFR exposuro nt reached ean d morem tham 0 n1 afte cycles4 r . This growt withis hrangwa e nth f otheeo r literature data.

The diameter of the characterized FRs changed very little, however, taking into account slight variations in claddings supplied by different vendors, as shown in Fig. 5.4. Thdiameterd ero s with claddings supplie vendoy db increaseK r d slightly with exposure. On the other hand, those supplied by vendor S changed only a little. However, when the contribution f oxid so unremovabld ean e crud were appeareconsidereds FR e o th dt l al , have decreased slightl diameteyn i creeo t e prdu downordee th f severay o r ,b l tenf so

~~5.3.3. Crud depositio wated nan r side corrosion~~

Some authors [5.4, 5.15] have reported no enhancement of the crud and/or external cladding oxidation of PWR Gd FRs. However, heavy nodular corrosion and significant oxide layers have been observed in BWRs on some Gd FRs although they have been irradiated in some cases at presumably lower powers than adjacent fuel rods [5.23]. A mechanism to explain this corrosion enhancement based on the ß'radiolysis of the primary wate bees ha rn proposed [5.24] summaryn I . , enhanced e corrosioth o t e ndu neutron absorbing material is explained by pair production from conversion energy capture photon claddine th n si g giving stronrisa o et g emissio f energetino . cß'

In conclusion, it is necessary to obtain more data on irradiated Gd FRs both in BWR poweR PW r d plantsan , relatin chemistre th wateo ge t th f yo r circuit, temperature th f eo cladding (i.e. LHGR frod direce man )th t in-pile environmeno f thesto o t FRsd s i eG t I . earl t presenya concludo t t e thapresence th t f absorbineo g material could inducy ean corrosion enhancement.

~~5.3.4. Reaction layer between claddinfued an l g material~~

In high LHGR FRs [5.2, 5.4], both Gd and U FRs have shown a pronounced interaction layer between claddin fued glan material. This reaction layer consiste f ZrOdo 2 (5//m caesiud an ) m uranate (10//m MWd/tM2 4 t a ) . Afte secona r d reactor cycl0 e(7 MWd/tM) caesium zirconate was observed. Such fuel cladding bonding appeared more indicatins FR importand G significan ga e th n i t t radial caesiu oxyged man n migration.

~~93 TABLE 5.3. PWR FRs IRRADIATION CONDITIONS~~

~~Programme MTR and/or %Gd Average Peak Peak Gas Refe- power plant burnup bumup LHGR release rences GWd/tM GWd/tM kW/m %~~

~~[5.1,5.2] BN- BR3 3 42 45 13.3 CEN/SCK [5.3, 5.4]~~

~~3 71 45 15.6~~

~~CEA Siloe Gd GRIF 5 1.75 20.5 2 [5.22]~~

~~8 7.4 29.9 12-17~~

~~CAP 3 15-16 25 25~~

~~3 25-29 37 25~~

~~10 12~~

~~BR3-R2 10 60~~

~~MHI- Halden 6 7 [5.19] (HBWR) NFI- 5 Utilities BR3 6 8~~

~~6 26~~

~~10 13~~

~~BR3-R2 10 40~~

~~DR3 10 3 days~~

~~NUPEC MIHAMA - 3 8.4 26 [5.9]~~

~~17.6 24~~

~~31.4 24 [5.19] OHI-2 6 2 124.- 6 16 < 1 [5.20]~~

~~94 TABLE 5.4. MAIN SPECIFICATION OF LEAD FAs IN FUKUSHIMA DAIICHÎ- 3 (BWR)~~

~~Item Specification~~

~~Lattice arrangement 8x8 square lattice Total length m 4.47~~

~~Fuel rod pitch mm 16.3~~

~~Fuel assembly Total heat transfer are2 m a 9.06~~

~~UO2 weight kg Abou0 t21 Total weight kg About 310 (including channel box)~~

~~Number 63~~

~~Effective length m 3.66~~

~~Fued ro l Plenum m length 0.4 s Fillga He Fill gas pressure kg / cm2 a 1~~

~~Diameter mm 10.6~~

~~m m Height 10.7~~

~~Pellet Material UO2 (partially UO2 - 2 w/o Gd2O3)~~

~~Density % TD 95~~

~~Number of water rods 1 Cladding Outer diameter mm 12.5 & m m Thickness 0.86 water tube Material Zircalo2 y- (recrystalized and annealed)~~

~~Number 7 Spacer Material Zircalo4 y- Inconel X - 750~~

~~95 Cladding tube average outer diamete) m m ( r Irradition growth (%AL/L) O O O p ro co en b) ro ro ro ro en 1 8 en g~~

~~0 t> O • [> 0 1*0 o c « N ; o o o o I N~~

~~•n «P' R en w » § ro jj) O 0) (Q~~

~~CO (D 3 C -Ö- 3 5- c TJ I——•-~~

~~l——e——i~~

~~M Q, l. P -e-~~

~~ro en - hOH~~

i A\ i I—'S?—I S At lower LHGR (20 - 30 kW/m), a thin interaction layer has been observed, although a trend towards an enhancement of the inner oxide layer has been reported [5.10] for Gd FRs. The thickness increase of the inner oxide layer is generally well correlated with the higher centre temperature of the Gd FRs resulting from the lowering of the thermal conductivity.

~~Furthermore, two observations are to be noted:~~

~~the relative mobility of Cs in the Gd FRs which axially migrated into the peak LHGR regio kW/m0 n(4 ) (see section 5.3.6.2.)~~

~~the formation of an oxidised compound containing Zr and Cs at the internal cladding surface.~~

~~Thes o observationetw explainee b n lowerina ca sy b d r stabilizatiogo e th f no chemical oxygen activity durin irradiatione gth .~~

~~It was observed after the first cycle in Gd FRs irradiated at high burnup and LHGR~~

in BR3 that the compounds present at the fuel-clad interface were ZrO2, Cs2UO356, and Cs2Te and after the second cycle, ZrO2, Cs2ZrOx, and Cs2Te. It can be deduced from the thermodynamic stability [5.25, 5.26] and from observations [5.25] that the following reaction occurred:

~~Zr0 + CsT ,Cs+ 2Tee 2 UO+ 3Cs.56 2UO Cs* - *i 2 CsZrZrOZ + x O + ZrTem + U0~~

~~AG(Cs2Te; j ) - AG(ZrTem)~~

~~AG(Cs2ZrOx) < •< AG(Cs2U03.56;~~

~~AG(Zr02) <: AG(U02).~~

chemicae Th l oxygen potential AGOsystee th n 2i m (Zr02, Cs2ZrOx, Cs2Te lowes )i r than the AG02 in the system (U02, Cs2U0356, Cs2Te). anothes i s FR rd indicatoG axia e e lowee Th th th l f n mobiliti o r s oxygeC f yo n activit volatilitdirectle s i ys Th . C prevailin f yo yFR relatee chemicas th it gn i do t l activity, whic higs hi h enoug induco ht e axial migration. Fro mthermodynamia c poin f viewo t , oxygen activity in the Gd fuel is not drastically modified in comparison with U or MOX fuel.

~~5.3.5. Microstructural fuel features~~

~~5.3.5. 1. Pellet density - densification - swelling~~

No significant differenc efue U betwee d l irradiatean d nG PWRn di s beeha s n geometricareportes a r fa s da l pellet stabilit concernedys i . Afte temporarra y densification durin firse gth t cycle, pellet density return as-manufacturee th o st e th f o d den value th t ea irradiation [5.9] whe LHGe nth arounRs i kW/m0 d2 . Pellet changes have been observed t highea r power ratings [5.2highee th ,o t 5.4r centre ]du e temperature FRse th f .so After restructuring, stabilit pellee th f yo t geometr reache s yi creep-dowo n d dan observes ni d between 48 and 70 GWd/tU.

In BWR FRs [5.9], the fuel density returns almost to the as-manufactured value at burnups of 30 GWd/tU. The densification appears higher for Gd fuel pellets than for U swelline pelletth t sbu g rat similas ei r (Fig. , IIIApp.20 . III).

~~97 5.3.5.2. Microstructure gadoliniumand diffusion~~

In FRs irradiated in PWRs at low LHGR (< 25 kW/m), no gadolinium diffusion has been observed and grain growth is also generally not observed at low burnup. At high burn-up, some incipient growing of the grains is observed in parallel with a small Gd diffusion [5.9, 5.15]. Above 25-30 GWd/tU burnup, large pores appear to develop in the

~~zone where GdO was previously in higher concentration. This phenomenon is 3~~

accompanied by 2 some grain growth. These pores, due to the difference between the U interdiffusiod G d an n coefficient vere ar ,y stabl contributd ean densite th o et y decrease of the gadolinia doped fuel [5.10].

~~At higher LHGR (> 30 kW/m), a central restructured zone is observed in which~~

~~complete solution of GdO in U0 fuel and grain growth are observed. In contrast to the 3 2~~

~~situation wit ratew hlo d FRs graie th , n 2 growth seem enhancede b o st completelo N . y satisfactory explanation has been given up to now [5.2, 5.4].~~

The grain growth of fuel irradiated in a BWR [5.5, 5.27] was measured at different radii during the PIE. The calculation based on the out-of-pile investigation overestimated the value in some cases. This was attributed to grain boundary bubbles observed in the corresponding specimens which retarded grain growth. The observations reveal that the grain centre pellee th th t f sa et o grow slightly during irradiation graie th ; n size differences between the centre and the periphery of the pellets are shown in Fig. 111.21 (App. III).

~~5.3.5.3. meltto Power~~

investigato T combinee eth d effec f reduceo t d thermal conductivit burnupd yan ,a power-to-melt experimental determination has been conducted on a PWR Gd FR in the e Internationaframth f o e l Programm (HjgC HB eh gurnup Chemistry) organizey b d BELGONUCLEAIRE. The results are restricted to the participants in the programme.

~~5.3.5.4. Conclusion on microstructure~~

When the LHGR is maintained around the nominal design value (15 - 20 kW/m), no significant modificatio initiae th f nlo fuel structur bees eha n observed. Some incipient diffusio gadoliniuf no m int fuee oth l matri bees xha n notecentr e hottese th th dn f i e o t fuel rods (20 kW/m < LHGR < 25 kW/m).

~~At high LHGR, no significant difference between Gd and U FRs has been recorded.~~

~~5.3.6. Fissio releass nfissioga d ean n product distribution~~

~~5.3.6.1. Fission re/easegas (FGR)~~

Both at high and low LHGR, the fission gas release from Gd and U FRs shows the same trend. No significant influence of the Gd presence has been recorded and the phenomeno believes ni more b eo dt relate fuee th l o dtemperaturt e gadolini e thath no t a solid solution in the fuel matrix.

Gd fuel irradiate peaa t da k rating belo kW/0 w2 m release smalsa l fractioe th f no total fission gas inventory [5.9, 5.10, 5.15]. Larger fission gas release rates seem to be observe hige th h n di burnu compares pFR burnudw [5.1lo wits e phFR th , 5.4].

98 Fission gas release data of FRs irradiated at high rating reveal similar release rates both for Gd and U fuel [5.2, 5.4, 5.22]. On the basis of the available results, gadolinia addition does not appear to affect the fission gas release rate. The most important factor for the gas release is therefore the centre temperature and its evolution with time [5.1, 5.4].

e puncturinTh 4 characterize4 f go powe R irradiates BW dFR r a plan n di t [5.9] showed the same trends for both Gd and U FRs. In the high burnup range, the FGR principally depends on LHGR (Fig. III.22 (App. Ill)) but less on burnup (Fig. III.23 (App. III)). The FGR of Gd fuel was found to be similar to that of U fuel.

~~5.3.6.2. Fission product distribution~~

In low-rate elements C d an dsd FRswitN , hhiga , Mo h fission yiel almose dar t uniformly distributed within the fuel pellet. Only some deposits containing Cs-U-O have been detected on the inner clad surfaces.

In contrast, high-rated FRs show an important radial Cs migration to the inner clad surface. Compounds containing Cs-U- Cs-Zr-d Oan O have e beeth f o n s observeFR e th dn i Belgian programme [5.1, 5.4]. Furthermore, axial Cs migration was observed in these Gd FRs and not observed in the U or MOX fuel rods irradiated in similar conditions (see section 5.3.4).

~~5.3.6.3. Conclusion on fission product behaviour~~

The fission product behaviou fuee th l n matrii r x doe t influenceseee sb no mo t y db presence solin i th d dG solutio f eo fuee th l n matrixi mose Th . t important parametes i r the temperature of the fuel matrix which governs the diffusion and transport processes.

~~5.4. BEHAVIOUR UNDER REACTIVITY-INITIATED ACCIDENTS~~

In-reactor experiments were conducted to study the failure behaviour of Gd FRs under reactivity-initiated accidents by using the NSRR. The specific objectives of the study were to determine the thresholds of fuel failure and mechanical energy generation an identifo dt fuefailure d yth ro l e mechanism compared wit fuelhU .

wor e performes Th kwa joina s da t study betwee I n[5.28] JAERMH d . an IDetail s are given in App. III. It was concluded that: failure Th e oxyge s mechanisi s . 1 FR nd induceG mf o d cladding embrittlement accompanied by wall thinning due to local melting of the cladding, which is same fuelth thaU s ea f .o t

~~cal/g.U0failur5 e 27 Th betwees d i e s thresholan 2. FR 5 , 2 almosd n26 G f do t the same as that of U fuel.~~

~~3. Both the Gd and U FRs have almost the same mechanical behaviour as a function of energy generation.~~

4. In consideration of the radial power profile in the fuel and the pulse power shape expecte LWR,n a failure n di th estimates i e R thresholLW n do t a dn i be at the same level or greater than the presented results. Indeed, in the

99 FRs of the NSRR experiment, Gd160-enriched Gd was used to provide high resulta reactivit s powea e ; tese ,th th FR t n ryi profil less ei swithiR F e nth depressed LWRthan a additionn n i I . ratie th ,f beta/energ oo y releases i larger in the NSRR core than in an LWR core and consequently the rates of power change (both power increase before and decrease after the peak) are faster than in LWRs.

~~REFERENCE CHAPTEO ST R5~~

[5.1] BAIRIOT, H., et al, "Gd fuel behaviour to high burn-up", Performance of fuel and Cladding Material under Reactor Operating Conditions, Proc. KTG Sem. Karlsruhe, 1985, Kerntechnische Gesellschaft Bonn (1985).

[5.2] BLANPAIN, P., HAAS, D., MOTTE, F., "High rated and high Burn-up Gadolinia Fuel irradiated in the BR3 17x17 PWR", IAEA-SM-288/40, Proceeding of Int. Symp. "Improvementn o Waten si r Reactor Fuel Technolog Utilisation"d yan , Stockholm, 15-19 Sept.1986.

~~[5.3] HAAS, D., MOTTE, F., "Experimental Programmes on Gadolinium Fuel Utilisation in LWRs in Belgium", Proc. Specialists Mtg Mol, 1984, IWGFPT/20, IAEA, Vienna, pp.171-181.~~

[5.4] BILLAUX, M., BLANPAIN, P., BOULANGER, D., "Evaluation of the Thermomechanical Properties of High Burn-up Gadolinia Fuel. ANS Topical Meeting "LWn o R Fuel Performance" Williamsburg, April 17-20, 1988. pp.356-363.

~~[5.5] BRUET, M., et al., "Analytical out-of-pile and in-pile experiments on gadolinium- bearin g, Geneva fuels"86 C EN ,, June 1986.~~

[5.6] BRUNA NOBILE, VALLEEd ,G. an . , "Transpor,M ,A. diffusiod an t n infinite medium assembly calculation powen pi f so r distribution", presente Internationat da l Meeting on Advances in Mathematics, computation and reactor physics ANS. Pittsburgh, PA, USA, April 1991.

~~[5.7] BONNIAUD, M., MAGNIN, D., NORSTROM, G., LARSSON, J., "Advanced Fuel Managements with Gadolinium Peripheral Design", Deutsches Atom Forum, Dusseldorf 1989y Ma , .~~

~~[5.8] SOHN, D.S., KIM, S.H., and YANG, C.K., "Korean Nuclear Fuels in the Nineties," Proceedings of the 6th KAIF/KNS Annual Conference, pp.529-539, April 1991, Seoul, Korea.~~

~~[5.9] NUPEC, "Proving Test on the Reliability for Nuclear Fuel Assemblies - Irradiation Tes Fuen to l Assembly", Summary Report Nuclear Power Engineering Test Center- 1987.~~

~~[5.10] MANZEL, R. and DOERR, W.O., "Manufacturing and Irradiation Experience With~~

~~U0/GdO Fuel", Ceramic Bulletin Vol. 59, No. 6 (1980), pp.601-616 (paper 2 3~~

~~presente2 d at the Annual ACS meeting, Cincinati, May '1979).~~

~~[5.11] SIEMEN SFue- l Assemblie s- Reference s '92.~~

~~100 [5.12] ANDERSSON, S.O., PATRAKKA , HYDENE. , , PETTERSSONL. , d an . H , GRAPENGIESSER , "ThB. , e Finnish-Swedish High Burnup Fuel Evaluation Programme", IAEA-SM-288/45, pp.281-289.~~

[5.13] SKOGEN, F.B. and KILLGORE, M.R., "Performance of Exxon Nuclear Gadolinia- Bearing Fue Pressurizen i l d Water Reactors" TopicaS AN , l FueMeetinR lLW n go Performance, Orlando, Florida, April 21-24 1985, pp.5-4 5-49o 3t .

[5.14] SKOGEN, F.B., O'LEARY, A.M KARCHERd an . , K.E., "Performanc f Gadolinieo a Burnable Absorber Assemblie t Burnupsa 50,00o t p su 0 MWD/MTU" TopicaS ,AN l FueMeetinR l LW Performance n go , Williamsburg, Virginia, April 17-20 1988, pp.348-355.

~~[5.15] YOKOTE t al.e , ,A Japanes" M. , e Gadolinia Fuel Usag d Demonstratioan e n Program", (ANS Trans. Oct. 1989).~~

~~[5.16] MATSUOKA Y., et al., "Gadolinia Fuel Development in Japanese PWR", Proceedings of an International Symposium in Tokyo by OECD, April 1989.~~

[5.17] BGUNA , NOBILE VALLEE,G. d an . , "TransporM , ,A. diffusiod an t n infinite medium assembly calculation n powepi f o rs distribution", International Meetinn o g Advances in Mathematics, computation and reactor physics ANS. Pittsburgh, USA, April 1991.

[5.18] POLJAKOV, A.A., et al., "Investigation of neutronic parameters of uranium water lattice wit absorberd hG Modellinn i " investigatiod gan f neutronino c processen si nuclear power reactors, Moscow, Energoatomizdat, 1991, pp.118-121.

~~[5.19] MORI K., et al., "Development of gadolinia Fuel for PWR", IAEA-SM-288/22P.~~

~~[5.20] YOKOT t al.e , , "GadoliniEM. a Fuel Usag Demonstratiod ean n Progra Japan"mn i , ANS/ENS FuemeetinR l LW Performance n go , Avignon Meeting, April 1991, p.897-908.~~

~~[5.21] DOI, S., "Post Irradiation Examination on the Gadolinia Containing Fuel Irradiation Domestie inth c PWR" Mitsubishi Genshiryok , 199014 . .pp u 7 Giho5 . uNo~~

~~[5.22] BRUET , PORROTM. , , E.,TROTABAS , MELIN BONNAUDd M. , an . P , "Analysi, ,E. s~~

~~of Thermomechanical behaviour of UO Gd0 Fuel Under Irradiation Conditions", 2 3~~

~~ANS Topical FueMeetinR l LW Performancen go , Orlando2 , Florida, April 21-24 1985, pp.5-5 5-63o 1t .~~

~~[5.23] MARLOWE, M.O., ARMIJO, J.S., CHENG, B., ADAMSON, R.B., "Nuclear Fuel Cladding Localized Corrosion" TopicaS AN , l FueMeetinR lLW Performancen go , Orlando, Florida, April 1985, p.3-73.~~

~~[5.24] LEMAIGNAN, C., "Impact of ß'radiolysis and Transient products on irradiation- enhanced corrosion of zirconium alloys", J. Nucl. Mat. 187 (1992) 122-130.~~

~~[5.25] DROUWART , PersonaJ. , l Communication V.U.B. Free Universit f Brusselsyo .~~

~~[5.26] RAND, M.H. and POTTER, P.E., Personal Communication UKAEA (AEA) HARWELL..~~

~~101 [5.27] KOGAI, T., et al., "In-Pile and Out of Pile Growth Behaviour of sintered U02 and (U, Gd) 02 Pellet", Journal of Nuclear Science and Technology, 26[8] p. 744-751, Aug. 89.~~

~~[5.28] SHIOZAWA t al.e , , "StudS. ,Behavioue th n yo f Gdo r 203 FailurFued Ro l e under a Reactivity Initiated Accident", (JAERI-M 88-084, May, 1988).~~

~~102 . 6 FUEL CYCLE BACD KEN~~

~~6.1. SPENT FUEL STORAGE~~

Since the discharge burnup of FAs with burnable absorbers is, in principle, identical to the burnup of standard FAs, no major difference is apparent between the two types of Fas. Amongst the more subtle differences, however, it should be noted that:

the power rating of Gd FAs at EOL is usually higher than the rating of the givea standarr fo n s reload e residuaFA d Th . l heat after shutdows i n therefore affected durin e initiath g l perio f speno d t fuel handlind an g storage. This difference disappears with time onc e residuath e l heat becomes predominantl poweL y affecteEO r y ratingb t burnuy db no .d pan

the relatively high Pu build-up and consumption during the first cycle due to the neutron spectrum hardening results in a slightly higher Pu content in the spent fuel and a shift of the Pu composition towards higher isotopes. This effect, limited to the Gd FRs in the FA, has a negligible impact on the spent FA characteristics.

~~6.2. TRANSPORTATION~~

~~The characteristics mentione n sectioi d 1 woul6. n d also influence spent fuel transportation. However, since transportation will only tak yeare0 1 plac o st aftee2 r EOL, the impact is negligible.~~

~~6.3. REPROCESSIN WASTED GAN S~~

G frequentlds i y addedissolvee th do t reprocessinn i r g plant increasso t margie eth n to criticality. Gd has been proven not to interfere with the chemistry of the process and to be recovered in the HLW stream to be vitrified.

e mosth tn I recent French reprocessing plants, criticality contro s achievei l d principall implementatioy yb f criticality-safno e geometr equipmente th f yo e th . t Thino s i dissolvee cas th UK' e r th efo s n i rTHOR P facility, where monitorin fissilr gfo e contenn tca be exploited to reduce the level of Gd addition in the process thereby reducing the volume of vitrified waste.

The main effect of the presence of Gd in irradiated fuel is to increase slightly the volum f vitrifieeo d wast consequentld ean cose f yth reprocessin o t wastd gan e disposal. The total oxide conten t affecteglase no th presence f s i th o t sinc d y doxidb e G ef th e o e loading is determined by the incorporation level set by plant operating parameters (althoug oxidd formw hG eno s e total)consequenc e th par f Th o .t f addino es i d gG potentiall reduco yt e quantiteth f Initiayo l Heavy Metal whose waste e oxideb n ca s accommodated in a single vitrified container.

The quantity of glass is determined by the total mass of waste oxide (including the inactive species) and its incorporation level. So far as activity is concerned, this will decay with time. At the time of vitrification the remaining active species constitute about 25% of the total fission product mass. The principal limitations imposed by activity levels are secondare th y effect f heaso t productio dosd nan e rate.

103 e quantitIth f f oxideincorporateyo e b o st glase th sn di limiting e werb o et e th , quantity of vitrified waste would only be increased by «7% for BWRs, « 1.5% for PWRs and «2.2 r WWERs%fo . Assuminreprocessine th f o vitrificatioe % gth 30 ge cosb o nt t [6.1] this would then increase the reprocessing cost by —2.2% for BWRs, =0.5% for PWRs and «0.7% for WWERs i.e. small enough not to be reflected in the reprocessing price.

~~REFERENC CHAPTEO ET R6~~

~~[6.1 ] OECD/NEA, "The economic nucleathe sof r fuel cycle", 1985.~~

104 . 7 CONCLUSIONS f burnablo e us ee absorbeTh r welfuea s i ll established technology resulting from over 30 years of commercial utilization in BWRs. Gadolinia has emerged as the most widely used burnable absorber. Applicatio PWRso nt , BWR WWERd san s together with trende th s toward increas f reactoeo r cycle lengt fued han l discharge burnup have resulted highen i contentd rG s being required, leadintechnicaw ne o gt l challenges. Experiencs eha shown that these challenges can be and have been met. It has, nevertheless, also fostered the development of two alternative solutions to Gd: zirconium diboride coating and erbium.

fuel Fod fuee G rth , l properties, manufacturing techniques, design approaches, utilization and performance as well as potential impacts on the backend of the fuel cycle ar l illustrateeal thidn i s repor providd tan e evidenc succes e maturite th th f eo d sthif an y o s type of fuel technology.

Gd resource adequate sar meeo et t expanded usag nuclean ei r knowfuele Th . n geological reserve establishee th d san d minin refinind gan g capacit copn yca e with even huge increases of demand from the nuclear industry, since this will never exceed a few percen annuae th f o t l productio other fo d r G purposes f no .

Many of the physical properties of Gd fuel are known with adequate precision. However larg,a e scatte observes ri reportedn i d measurement thermaf so l conductivitd yan meltin gpropertiey pointke o ,tw s affecte presence th . Thiy d b Gd probabl s i f eo y due, in part, to the use of different fuel manufacturing techniques.

The neutronic propertie wele sar l known t theibu r incorporation into nuclear design methodology requires further attention overlappino t e du , f certaigo n resonanced G f so with those of U and Pu isotopes. The evolution of radial power distribution across the pellet is very different in Gd fuel compared with standard U fuel and requires a specific adaptatio fuee modellind th ro lf no g code.

considerable Th e numbe alternativef ro positioninr sfo fued G l axiall e fuege th th l n yi rod, radiall fuee th l n yassembli radialld core bees yan th eha n y i utilize y fuedb l vendors to meet various objectives. The flexibility this affords has resulted in quite different approaches, some of which are described in the report.

performance Th fuebees d G ha l n f e o goo post-irradiatiod dan n examinations have provided a database to validate the design methodology. This data base is constantly being augmented and updated to provide a better statistical validation and to cover higher performance regimes.

~~The backend of the fuel cycle is practically unaffected by the addition of Gd to the fuel.~~

It should be noted that this report represents achievements up to the period 1991- 93. After a few more years, it would be useful to update the status and to compare Gd fuel with its alternatives, which by then will have achieved a greater number of reactor- years of operating experience, enabling a more useful statistical comparison to be made.

~~105 Appendix I GADOLINIUM AND ERBIUM - A REVIEW OF RESOURCES AND SUPPLY~~

~~E. Müller-Kahle Division of Nuclear Fuel and Waste Management, International Atomic Energy Agency, Vienna~~

~~. 1 Introduction~~

Gadolinium (Gd) and Erbium (Er) belong to the rare earth elements (REE). The name rare earth reflects the earthy appearance of the discovered oxides and the perception of the early discoverers that these elements were present in the earth's crust only in small quantities.

A number of REEs were discovered in geological material from the town of Ytterby, near Stockholm, Swede Kary nb l Axel Arrheniu n 178si 7 [1]yeae .On r later, Bengt Reinhold Geifer further investigated the rocks of Ytterby, which later were named gadolinite after Johann Gadoli Finlandf no liveo d,wh between 176 0- 1852 t seem.I s that Gadoli separated nha frod dG m this material, althoug thoughe hh found t thaha e dth e h t element yttrium. Fro e sammth e material that Gadoli discarded ha n n 1794di , G.G. Mosander in 1843 discovered Er together with yttrium and terbium.

Gd and Er belong to the lanthanide series, which together with the elements scandiu yttriud man m constitut havr E atomie ed th REEsan e ecth d numberG . d an 4 s6 68 respectively. Gd belongs to the REE group with high melting and boiling points, while Er also has a high melting point but a low boiling point. Both elements belong to the "heavy" REEs, also yttriue referreth s a m o dt subgroup .

~~2. Industrial application of REEs~~

~~The REEs have a wide range of industrial applications in a number of sectors. An overvie followingivee ws i th n i g table [1].~~

~~Metallurgy: alloying elements in iron, steel, superalloys, etc., and for corrosion resistant materials.~~

~~Glass: polishing compounds, radiation stabilizer, ingredient to increase refraction etc...~~

~~Ceramics: high temperature refractories, engines, metal substitutions, coatings.~~

~~Electronics: cathodes, electrodes, semiconductors, computer memorie substratesd san , lasers,~~

~~Chemicals: l catalystoi refineries r fo s , chemical processin d analysesan g , pharmaceuticals.~~

~~Magnets: electric motors, alternators, generators, computer disk drive actuators, proton linear accelerators, headphones, speakers.~~

~~Nuclear: control rods in nuclear submarines, burnable absorbers, shielding material, reprocessing additive,~~

~~107 TABLE 1.1. WORLD REPORTED RESERVES OF RARE EARTHS ( TONNES RAREOF EARTH OXIDE)~~

~~Country Reserves~~

~~North America U. S. A. 5 500 000~~

~~Canada 164000~~

~~South America Brazil 20000~~

~~Europe Finland, Norway, Sweden 450 000~~

~~Former USSR 50000~~

~~Africa Burundi 1 000~~

~~Egypt 100000~~

~~Kenya 13000~~

~~Madagascar 50000~~

~~Malawi 297 000~~

~~South Africa 350 700~~

~~Asia China 36 000 000~~

~~India 1 800 000~~

~~South Korea 45000~~

~~Malaysia 30000~~

~~Sri Lanka 13000~~

~~Thailand 1 000~~

~~Oceania Australia 345000~~

~~World Total (rounded) 0 00 6 23 5 4~~

~~Source . BureaS . U : f Mineuo s~~

~~108 Other: jewellery dryersk , painin d t,an hydrogen fuel storage, refrigerant, lubricants, thermometers.~~

~~. 3 Resources~~

~~The available literature refer genera n i swhole groupth E f o availableI t eRE l . , special emphasi thin i r s E paperd s an wil placee d . b lG n do~~

~~The REEs are widely distributed in a variety of geological environments. The overall abundance of the REEs in the earth's crust is estimated to be of the order of 150 - 200 parmillior pe t n (ppm).~~

concentrationr E d Gdan igneousn i s rock differenf so t chemical compositions show only small variations: the Gd traces in igneous rocks ranging from basaltic to granitic composition varppm4 7. y o ,t fro wit 2 averag mthesn r h6. a ppm 2 E r 7. e f Fo e.rangeo s are 3.6 to 4.2 ppm, with an average of 3.9 ppm [2]. Higher concentrations are found in alkaline rocks e.g. syenites carbonatitesd ,an thesr Fo .e roc kcontentd typeG e sth s range from 8.7 to 79 ppm and those for Er from 3.6 to 48 ppm [2].

Mineralogically, there are a large number of minerals containing REEs. However, no mineral could be found in the literature where Gd is included in the chemical formula, while Er is contained in the chemical formula for the mineral fergusonite (Y, Er, U, Th) (Nb,

~~Ta, Ti) 04.~~

~~economicalle Th y importan mineralE RE t s are:~~

~~Minera l contenE RE tComposition~~

bastnaesite: CeFC03 max. 75 % REO *

~~monazite: (Ce, Y) PO2 65 % REO~~

~~apatite: (Ca,Ce)5/(P,Si)O4/3(0,F) 12% REO~~

~~brannerite: (U,Ca,Fe,Y,Th)O RE 3 (Ti,Si)% 0 7 506~~

~~gadolinite: (Y,Ce)2FeBe2Si2O10 48 % REO~~

~~O RE % 2 6 xenotime: YP04~~

* REO = Rare earth oxides

~~t seemI probabld san , thaGd ytcontaine s i als , oEr thesn di e minerals replacing some of the other REEs in the crystal lattices.~~

The worldwide resources of REEs, as reported and published [3], are mainly located summarA . ) knowe th % f yo 2 n (1 A US e th Chin n n i i totale d th af an ) o (nearl % 0 y8 resources is given in Table 1.1.

Of these resources vase th , t majorit estimates i % f ovelocate yo e 0 b 9 r o dt n di carbonatites and associated rocks, and the remainder in placers, beach sands, etc. In additio o thesnt e known resources whicn o , h reliable informatio availabls ni e th n i e literature, ther othee ear r "REE resources which canno classifiee b t knows da n resources, but rather as speculative resources.

109 These include, for example, a total of 100 million tons RED in China as well as the Mount Weld deposit in Western Australia with resources of 6.2 million tons REO. Also, there are a number of deposits, associated with the East African rift zone, which are not yet fully explored. Furthermore, there is a potential for REE in the large alkaline complex of liimaussa Greenlandqn i .

~~These examples show that the potential to increase the known resource base listed above, is good.~~

~~4. REE Production~~

~~productioE RE e Th n comes essentially fro same mth e countries which hose th t resources liste Tabldn i e 1.1.~~

~~latese Th t dat198e ] refeth a[3 o 9t t rREO productio0 00 , 0 distribute5 f no s da follows:~~

~~t RE 0 O20 7 Australia:~~

~~Brazil: 1 100 t REO~~

~~Canada: 100 t REO~~

~~t RE 0 O00 0 2 China:~~

~~India: 2 200 t REO~~

~~t RE 0 O30 3 Malaysia:~~

~~t RE 0 O80 Thailand:~~

~~USA: 14 000 t REO~~

~~t RE forme0 O50 1 r USSR:~~

~~t RE 0 O10 Others:~~

Past REE production available sar e fro sourceo mtw differenr sfo t period aread san s (WOCA/non-WOCA): 1981 and 1982 for WOCA [4], 1983 and 1984 for WOCA + non WOC same = Aworlth y edb r author198fo d 6 worle - an 199, th r [3]n e di 0fo Th . respective dat r thesafo e year aread san s are:

~~1981 28 535 t REO (WOCA only) t RE 0 O34 (WOC8 2 A only)1982~~

~~1983 36 000 t REO (world) t RE 0 O00 (world0 5 ) 1984~~

~~1985 not available t RE 0 O75 (world8 3 ) 1986~~

~~110 (worldO RE t )0 47 4 4 1987 1988 46 000 t REO (world)~~

~~(worldO RE t )0 00 0 5 1989~~

The most important REE production centres in hard rock environments are Molycorp's bastnaesite Mountain Pass Baotoe minth Californian ei d uan minA t eUS a , Bayan Obo, Inner Mongolia, China. Both producers account for over two thirds of the 1989 world production smalA . l amoun f REEo t , mostly yttriu mretrieves i byproduca s da t from U mining in Canada's Elliot Lake district. As U mining in this area is ceasing, it is expected tha yttriue th t m recovery will als terminatede ob .

The remainder of the REE production is issued from beach placer deposits in Australia (Eas Wesd an t t coasts), Brazil, India Soutd an , h Africa, althoug Palabore hth a mine recovers monazite from carbonatite productione Th . Malaysin i Thailand aan d comes fro mininmn ti g tailings consistd an , s mainl f monazityo xenotimed ean .

Specifi t availableno e c ar production orde I r E . o provid t rd an d ne G dat r fo a reasonabl contenr E mosee d th estimates f tan o t importand G e th , t source minerale sar given below expresse [5] f totaO o : RE % l s da

~~Gd Er~~

~~bastnaesite 0.35 0.01~~

~~monazite 5.00 0.22~~

~~xenotime 4.00 5.40~~

~~The 1986 productio s estimatei n o includt d t bastnaesite e 0 abou00 2 2 t , 30 000 t monazite and 250 t xenotime. Using the above contents, this production amounts to about 1 300 t Gd and 75 t Er.~~

Using the average Gd and Er contents of over 3.5 million t REO evaluated in [4] in 1986 and applying them to the total production of 50 000 t REO reported at the beginning of this chapter, the Gd and Er contents are estimated as 500 t and 50 t respectively. Based on these two approaches, it is estimated that the total Gd and Er respectivelya t/ 5 7 - 0 5 d . an a t/ production 0 30 1 abou- e s0 ar 50 t

~~5. Suppliers of REEs~~

The suppliers of REEs in their different forms i.e. concentrates or REE as metals can be divided into three groups: suppliers of concentrates, vertically integrated concentrate producers with subsequent processing facilities and finally the processors, which process purchased concentrates.

The most important suppliers of concentrates are Allied Eneabba Ltd. and Associated Minerals Consolidated Ltd., which mine beach sand deposits in Australia. Other, less important, producer locatee sar Thailandn di i LankaSr , , South Africa, etc.

~~The Molycorp., Inc. operato e mountaith f o r n Pass min n Californii e d an a processing facilities in Washington and York, Pennsylvania, and the Baotou Iron & Steel~~

111 Colargese Chinn i th . e aar t integrated REE suppliers. Als Chinon Ganije i th s ai a Rare Earths Corp., a partnership between the Canadian Pacific Rare Earths & Metal Corp. and theGanzhou Metallurgical Industries Corp., which produces concentrates and processes the mREOo t 198 e Th . 7 productio reportes nincludwa o t ] d[5 e oxides d nearlG t y5 . Japann I o processor, tw ther e ear s with foreig materiaw ra n l sources: Mitsubishi Chemical Industries Ltd., which has a joint venture with the Malaysian producer Beh Minerals, and Shin-Etsu Chemical Co. Ltd., which participates (or participated) with Denison and Molycorp in the recovery of yttrium from the Elliot Lake uranium mine in Ontario, Canada. Other integrated supplier Brazin i e sar l (Nuclemon Indid an )a (Indian Rare Earths Ltd.) which recovers Th in addition to the REE, including Gd.

The most significant REE processor is Rhône-Poulenc, France, with worldwide interests mais It . n plant locatee Rochellesar a L dn i , Franc Freeport,d ean Texas, USA. In addition there are a number of Japanese processors which import raw material mainly from China.

~~6. Market~~

The REE market is experiencing profound changes due to the declining demand certain i n "traditional deman"e areath d san d maio growt fieldsw tw ne ne areahn .Th i s demanE RE whers decrease ha de e petroleu th th ee metallurgica e th ar dd man l industries. In the petroleum industry, REEs are mainly used as catalysts for the cracking of leaded gasoline, which in many countries is being replaced by unleaded gasoline, steee whil th f hig o n el i industr he strengtus e yth h steel r pipeline l sfo drillin oi d gsan pipes has decreased over the past years.

In a broad sense, there is a compensation for the above decline, through the increase in REE demand in new fields such as magnets (neodymium, dysprosium), automotive catalysts (yttrium, cerium), additive r phosphorfo s d coloun an i s V T r computer screens (yttrium and gadolinium) and superconductors (1-2-3 compounds: 1 atom REE, 2 atoms barium and three atoms of copper).

~~maie Th n current marketd an t A specifiedsno (yeaUS s e wa r th REE fo )e ar s Japan shows .A n below each RE differens Eha t uses resul a [3] demans e .A tth d differs between REEs:~~

~~USA: Petroleum catalysts 53% of demand~~

~~metallurgical additives 22% "~~

~~ceramics and glass 18%~~

~~electronics, phosphors, % 7 magnets etc.~~

~~Japan: f demano % glas 70 ceramicdd san s~~

~~magnets 19% "~~

~~phosphors 6%~~

~~petroleum catalyst & metallurgical additives 5%~~

~~112 These different trends are also reflected in the demand data for the USA and Japan availabl perioe th r defo 198 3- 198 7 [5]:~~

~~1983 1984 1985~~

~~USA1 196002 21400 12100 11800 9400~~

~~Japan 3870 4840 5450 7105 5673 5~~

~~1 Estimate 2 Tonnes REO~~

Includes an estimated 850 t materials for RE magnets. 3 Details on the composition of the above demand data are not available. However, judging from the end uses, it can be assumed that the US demand is mainly for the "light" REEs, comprising the first seven elements of the series (lanthanum - europium), whil Japanese eth e deman mainlys e "heavydi th (gadoliniu E r fo RE " m- lutetium plus yttrium), whose projected growth rat estimates e i p/year % 5 1 .t da

REEs including Gd (no information on Er) have been traded since 1988 on the Industrial Metal Exchange (IME). Recent prices for Gd or Er are not available. An indicatio approximate th f no pricreferencd a eG s i e [5]n ei , wher price eth e rangr efo 99.99% pur frod eG m Molycorp's refinerie quotes i 65.0s da 0- 70.0 0 USD/lb.

~~REFERENCES TO APPENDIX I~~

~~[1] MORRAL, F.R., A history of the rare earth elements, CIM Bulletin, November 1990.~~

~~[2] WEDEPOHL, H., Handbook of Geochemistry, Springer, Heidelberg, 1978.~~

~~[3] SPOONER, J., GRACE, K.A., ROBJOHNS, N., The economics of the rare earth elements BulletinM CI , , 1991.~~

] [4 ANSTETT, T.F., Availabilit f raryo e earth, yttrium related ,an d thorium oxides- market economy countries minerala , s availability appraisal; U.S. Bureaf uo Mines, Information Circular 9111, Washington, D.C. 1986.

~~] [5 O'DRISCOLL , RarM. , e earth, Ente dragone th r , Industrial Minerals, November 1988.~~

~~113 AppendiI xI GADOLINIA FUEL UTILIZATION IN FRANCE~~

~~This Appendix provides further detail f Frenco s h experience wit fued s hG a l summarize Chaptedn i . 5 r~~

Table 11.1 gives details of reload characteristics (cycle lengths, numbers of GD FRs etc.) for all plants for which FRAGEMA has supplied Gd fuel, as summarized in Section 5.1.2. Figure 11.1 provides details of the core loading patterns for IN-OUT fuel management with Gd FRs in peripheral locations.

Table II.2 provides detaile resultth f f comparisono o s f predictiono s s with measurements of core physics parameters. These are referred to in Section 5.2.2.1. Figure II.2 provides a graphical representation of measured and predicted FA power evolutions, as also mentioned in Section 5.2.2.1.

~~TABL FRAGEMA'L EIL S GADOLINIUM EXPERIENC POWEN EI R REACTORS~~

~~Date Cycle leigtfc (hl,- "i«, of cycle) RdMd's (MWd/t) NoBber j -J „^ ^__B^J_ J Phot/cycle CBBTXQiBiSDC_•__- — „J- —tmlimm -: S SDWH ^IC^HKB •fU/Gdrads ofU/Gdrods~~

~~1983 Gravetines 2 52 assemblies Cycle 3 {3.2E= 5 w/o) (8 Gd-beariiig 11020 mid-assembly 96 assemblies) 1984~~

~~GraveUnes 2 64 assemblies Cycle 4 (E = 3.45 w/o) 13820 mid-assembly 288 (36 Gd-bearmg assemblies) Tricastin3 64 assemblies Cycle 4 (E = 3.4f w/o) 15050 mid-assembly 288 (24 Go-bearing assemblies)~~

~~GraveUnes 5 1st core Cycle 1 (24 Gd-bearing 14925 mid-assembly 352 assemblies)~~

1985 Graveunes 2 64 assemblies Cycles (E = 3.45 w/o) 15330 uiid-asseiiibly 288 (36 Gd-beariug assemblies) 1986 Tricastin3 52 assemblies Cycle 5 (E = 3.70 w/o) 12035 mid-assembly 128 (16 Gd-beariug assemblies)

~~114 TABL l (cont.. EII )~~

~~1986~~

~~Gravetines 1 52 assemblies Cycle 6 (E = 3.70 w/o) 13590 mid-assembly 128 (16 Gd-bearing assemblies) Chinol nB 52 assemblies Cycle 4 (E3.7= 0 w/o) 13360 peripheral 128 (16 Gd-bearing assemblies)~~

~~Ringhalc 4 32 assemblies Cycle 4 (E = 3.40 w/o) (8 Gd-bearing assemblies) 10380 peripheral 96~~

~~8 assemblies (E3.1= 0 w/o) 1987~~

~~Chinol nB 52 assemblies CycleS (E3.7= 0 w/o) 14055 mid-assembly 128 (16 Gd-bearing assemblies)~~

~~Ringhalc 4 48 assemblies Cycle 5 (E3.4= 0 w/o) 11920 peripheral 128 (16 Gd-beariug assemblies)~~

~~Tihange 2 48 assemblies CycleS (E = 3.80 w/o) 12390 mid-assembly 128 (16 Gd-bearing assemblies)~~

~~1988 Gravelines 1 52 assemblies Cycle 7 (E = 3.70 w/o) 13890 mid-assembly 128 (16 Gd-bearing assemblies)~~

~~Ringhals 4 32 assemblies Cycle 6 (E = 3.50 w/o) (20 Gd-beariug assemblies) 9508 peripheral 160~~

~~8 assemblies (3.4E= 0 w/o)~~

~~Tihange 2 52 assemblies Cycle 6 (E = 3.80 w/o) 13800 mid-assembly 128 (16 Gd-bearing assemblies)~~

~~Tihaiige3 52 assemblies Cycle 4 (E = 3.80 w/o) 11800 mid-assembly 128 (16 Gd-beariug assemblies)~~

~~115 TABL l (cont.. EII )~~

~~1989~~

~~Ringhak 4 44 assemblies Cycle? (E = 3.50 w/o) 11700 peripheral 192 (24 Gd-bearing assemblies)~~

RinghalsS 32 assemblies Cycle? (E = 3.50 w/o) 9950 peripheral 96 (12 Gd-bearing assemblies) Tihange2 52 assemblies Cycle? (E = 3.80 w/o) 13510 mid-assembly 128 (12 Gd-bearing assemblies) Tihange3 52 assemblies CycleS (E = 3.80 w/o) 14190 mid-assembly 128 (16 Gd-bearing assemblies)

~~1990~~

~~Ringhals 4 37 assemblies CycleS (E3.5= 0 w/o) 11860 peripheral 128 Gd-bearin6 (1 g assemblies) Ringhals 3 32 assemblies CycleS (E3.5= 0 w/o) 10240 peripheral 64 (8 Gd-bearing assemblies)~~

~~Tihange2 52 assemblies CycleS (E3.5= 0 w/o) 11025* mid-assembly 128 (16 Gd-bearing assemblies)~~

~~Tibange 3 52 assemblies Cycle 6 (E3.8= 0 w/o) 12500* mid-assembly 128 (16 Gd-bearing assemblies)~~

* natural cycle length

~~116 TABL (cont.1 L EI )~~

~~1991~~

~~Ringhals 4 48 assemblies Cycle 9 (3.5E= 0 w/o) 11430 peripheral 224 (28 Gd-bearing assemblies)~~

~~Ringhals 3 36 assemblies Cycle 9 (E = 3.50 w/o) 10310 peripheral 96 (12 Gd-bearing assemblies) Tihange2 52 assemblies Cycle 9 (E = 3.80 w/o) 15250 mid-assembly 128 (16 Gd-bearing assemblies)~~

~~Tihange3 52 assemblies Cycle? (3.8E= 0 w/o) mid-assembly 128 (16 Gd-bearing assemblies)~~

~~1992 KKP2 4 demo Cycle8 assemblies 11300 mid-assembly 32 (E3.4= 8 w/o)~~

~~KBR 8 demo Cycle 6 assemblies 64 (E = 3.95 w/o) Uranium matrix mid-assembly 2.3 w/o axially truncated Gdrods~~

~~Ringhals 3 32 assemblies Cycle 10 (E = 3.60 w/o) (20 Gd-bearing assemblies) 12 assemblies 11372 peripheral 192 (E = 3.50 w/o) (4 Gd-bearing assemblies)~~

~~Ringhak 4 28 assemblies Cycl0 e1 (E = 3.60 w/o) (20 Gd-beariug assemblies) 11122 peripheral 160 12 assemblies (3.5E= 0 w/o)~~

~~Tihange2 52 assemblies Cycle 10 (3.8E= 0 w/o) peripheral 128 (16 Gd-bearing assemblies)~~

~~1993 Ringbals3 36 assemblies Cycle 11 (E = 3.60 w/o) (16 Gd-bearing assemblies) peripheral 128~~

~~Ringhals 4 36 assemblies Cycl1 e1 (E = 3.60 w/o) peripheral 128 (16 Gd-bearing assemblies)~~

~~KKP 4 assemblies Cycle 9 (E = 3.48 w/o) Uranium matrix mid-assembly 32 o w/ 3 2. Axially truncated Gd rods~~

~~117 TABLE II.2. RINGHALS 4 — CYCLE 11~~

~~CORE PHYSICS TEST RESULTS~~

~~CONTROD LRO RESULTS DIFFERENCE CRITERIA PARAMETER CONFIGURATION (C-M) MEASUREMENT CALCULATION~~

~~Critical boron ARO 1496 ppm 1499 ppm + 3 ppm 50 ppm concentrations Din 1360 ppm 1359 ppm -1 ppm (BOL - HZP) D + Cin 1188 ppm 1191 ppm m pp 3 +~~

~~Isothermal temperature ARO - 10.5 fcmJ'C - 9.5 poM/*C + 1.0 PCB/'C 5.4 pmrC coefficient (BO LHZP- )~~

~~CYCLE LENGTH COMPARISONS~~

~~NATURAL CYCLE LENGTH AT 0 PPM (MWd/tU) DIFFERENCEC ( -M) CYCLE (MWd/tU) MEASUREMENT CALCULATION~~

~~RINGHALS 4 CYCLE 10 9976 10164 + 188 RINGHALS 3 CYCL0 1 E 10459 10561 + 102 CYCLE 11 9523 9729 + 206~~

~~118 12345678 12345678~~

~~CYCLE? CYCLES~~

~~12345678 12345678~~

~~1 1 [öd] 2 2~~

~~3 Gd 3 4 4 5 5 6 6 7 7~~

~~8 8 CYCLE9 CYCLE 10~~

~~12345678~~

~~1 Gd 2~~

~~3 Gd 4 FRESH ASSEMBLY 5 Gd DEPLETED ASSEMBLY 6 Gd 7 8 CYCL1 E1~~

~~Fig. 11.1. Ringhals Core— 4 loading patterns with fresh assembly locations~~

~~119 -O '4 . l~~

~~A 1 . 35 sS E M B L Y P O W E R - O 3 . l~~

~~25-t. l , 2OOO 4OOO 6OOO 8OOO 1OOOO 12OOO CYCLE BURNUP (MWd/tU)~~

~~••4O l.~~

~~A l . 35 S S E M B L Y P O W E R - 30 l.~~

~~1 . 25H 2OOO 4OOO 6OOO 8OOO 1OOOO 12OOO CYCLE BURNUP (MWd/tU)~~

~~Fig. 11.2. Ringhals - cycle 4 . Powers 9 FA D evolutionG o tw f o~~

~~120 Appendix III GADOLINIA FUEL UTILIZATIO JAPANN I N~~

~~This Appendix provides further details of the experience with Gd in Japan, as summarized in Chapter 5.~~

~~fue111.1d G l .developmen utilizatiod tan n~~

~~Table 111.1 shows the historical development of Japanese BWR design from 7x7 designJ B 8 presene x , givinth 8 y o parametert y da tgke f eacso h design. Thi referres i d Section i o t n 5.1.4.~~

~~Figures III. 1-2 show the geometrical arrangement of the latest 8x8 BWR assembly design; Figure III.3 show geometricae sth designR PW l .7 arrangemen1 x 7 1 a f o t~~

Table III.2 give breakdowsa f tesno t reactor irradiation detail r Japanessfo R ePW Gd FRs. Table 111.3 provides corresponding details for the commercial irradiation of Gd FRs botn i ans h dPWRFA BWRsd san . Thes referree e ar Sectio n i o dt n 5.1.4.

~~Tables III.4-6 give typical fuel loading scheme loop4 r loo 3 s loofo 2 , d p pPWRan s respectively.~~

Typical core loading pattern equilibriue th r sfo m cycl Japanesa f eo loope4 loo2 , p loo3 d p an plant showe sar Figuren i s 111.4, III.111.1d 8an 2 respectively. Corresponding critical boron curve showe sar Figuren i s 111.5, III.111.1d 9an 3 respectively, radial power distributions are shown in Figures III.6, 111.10 and 111.14 respectively, and corresponding burnus graphv ^ showe F pf ar s o Figurenn i s III.7111.1d an , 5 1 III respectively1 . . These ear referre Section i o dt n 5.1.4.

Startup test measurements comparisond an , s with prediction fued G l r loadesfo d in Mihama and Ohi (Table lll.7)are given in Table III.8. Figures 111.16 and 111.17 provide a graphical representatio comparisone th f no f criticaso l boron concentratio assembld nan y power. These are referred to in Section 5.2.2.2.

~~Figures 111.1 d 111.1an 8 9 give axial gamma scannin d radiaan g l micro gamma scanning plots respectively t celobserves a ,ho l examinatione th dn i s referre Section i do t n 5.2.3.~~

~~PIE results, covered in Section 5.3, are shown in Figures III.20 - III.23.~~

~~III.2. Safety tests~~

The behaviour under reactivity initiated accidents, correspondine th o t g experimental test matrix carried out in the NSRR and the test results, summarized in Section 5.4detailee ar , d further below:

~~Test Fuel~~

wers ThreFR eexperimene e R use th type r PW d fo f so shows a t Tabln i e III.d 9an Fig. III.24 fuee Th l. desig same th tha s ns ei a f commercia o t fueR l excepPW l r fo t lengthe th fuee ,th lisotopi d enrichmenG e cth d compositionan t .

~~121 Irradiation Condition~~

NSR swimmina s Ri g pool type thermal neutron pulse reactor core Th e. consists of 149 fuel rods, 3 transient rods, 6 regulating rods and 2 safety rods as shown in Fig. III.25. The pulse power is generated by rapidly withdrawing transient rods. The maximum available reactivity is 4.7$. Fig. III.26 shows a typical shape of pulse powe f 4.67o r $ reactivity, which generates 2110 peaW 0M k power7 11 , integrates MW d powe 4.4md an r s pulse width.

~~Test Results~~

Table 111.10 summarize e tessth t results. Typical fuel failure showe sar Fign ni . III.27. Fuefailurd ro l e thresholds derived fro resulte mth showe sar Tablnn i e 111.11. The cladding deformation as a function of energy deposition is given in Fig. III.28.

~~Table m.1 Typical BWR Fuel Design~~

~~Fuel Type 7X7 Adv. 7X7 8X8 8 X 8 RJ J 8B X 8~~

~~Wax. linear heat rate (kW/m) 57.4 60.7 44.0 44.0 44.0~~

~~Avg. Discharge Burnup (MWd/t) 21,500 27, 500 0 50 . 27 29, 500 33, 000~~

~~Pellet~~

~~Material UOS UO, U02 U03 UO,~~

~~U0,-Gd303 UOj-GdjOj UOj-GdjO, U02-Gd503 Diameter (mm) 12.4 12.1 10.6 10.3 10.3 Height (urn) 22 12 11 11 11 Stack Length (rrm) 3.660 3.660 3.710 3.710 3,710~~

~~Cladding Material Zr-2 2r-2 Zr-2 Zr-2 Zr lined Zr-2 Diam (inn) 14.3 14.3 12.5 12.3 12.3 Thickness (mm) 0.81 0.94 0.86 0.86 0.86 Liner Tickness (mm) —— —— —— —— ~0. 1~~

~~Fuel Assembly Numbe f Fueo r l Rods 49 49 63 62 62 Numbe f Wateo r r Rods 0 0 1 2 2~~

~~Beginning of Cocmercial Use Early 1970 's Mid of 1990 's En f do 197 0s ' Early 1980s ' End of 1980 's~~

~~122 Table ffl.2 Test Reactor Irradiation (As of mid.'91)~~

~~GdzOs con- Numbef o r Irradiation Burn-up Fuel Rod Reactor Name Type PIE Power Ranp Note tent (wt*) Gd Fuel Rods Period ) (Fav R Fabricator~~

~~HBTO 17x17 NOT (Halden short 6 1 4 Feb~'8'8 g Au 5 7GWd/t Mitsubishi None Reactor) rod DT~~

~~HBtfR 17X17 NDT (Halden short 1 0 1 90 Jun- lOGWt Mitsubishi None Reactor) rod DT~~

~~17x17 '84 Jul- '85 Nov 10GWd/t NDT BR-3 short 6 4 Mitsubishi One rod GAIN rod n 4 Ju Jul '8 7 ~8 27GWd/t DT~~

~~17X17 NDT BR-3 short 1 0 2 n 6 JulJu '8 7 ~8 12GWd/t Mitsubishi Ond ro e rod DT~~

~~BR-3 17X17 '86 n JulJu 7 -8 (NDT) 1 short 1 0 1 ~60GWt Mitsubishi —— R-2 rod 9 Aug'8 ~ (DT)~~

~~17x17 NDT NSRR short 6 1 0 p 3 Se Dec '8 4 -'8 unirradiated Mitsubishi 10 rods rod DT~~

~~17X17 NDT HBTO short 6 1 '84 Feb~'85 Aug 7GWt NFI NONE rod DT~~

~~17X17 v 4 No Jul '8 5 ~8 8GWd/t NDT BR-3 short 6 4 NFI ONE ROD GAIN rod '84 Jul- '87 Jun 26GHd/t DT~~

~~17x17 NDT BR-3 short 1 0 4 '86 Jul- '87 Jun 13GWd/t NFI D RO E ON rod DT~~

~~BR-3 17X17 (NDT) 1 short 1 0 1 '89 Apr- ~40G«d/t NFI —— R-2 rod (DT)~~

~~BR-3 17X17 1 short 1 0 1 '90 Sept (3day) —— NFI NDT —— DR-3 rod~~

~~123 Tabl3 . u l eCommercia l Reactor Irradiatio f mid.'91o s (A n )~~

~~Max. Burnup Disharged and In Use Irradiation (GWd/t) Number of FAs Number of Gd Rods~~

~~PWR 14x14 1988-up to date Mitsubishi 24 16 192 10ft~~

~~PWR 14X14 1990-up to date Mitsubishi 6 8 96 12ft~~

~~PWR 15x15 1989-u o datt p e Mitsubishi 17 52 832~~

~~PWR 17x17 1984-up to date Mitsubishi 33 224' 3584~~

~~PWR 14X14 1989-up to date NFI 19 28 336~~

~~PWR 15X15 1991-up to date NFI 3 28 448~~

~~PWR 17x17 1984-up to date NFI 33 100' 1600~~

~~8 Ptf8x R 1978-u o datt p e NFI 38 2768 19480~~

* Including 4 LTA,

~~Max. Burnup Discharged and In-Use Fuel Type Irradiation Supplier (G«d/t) No f Ass.o y No. of Gd Rods~~

~~BflR 7x7 1975—1982 Hitachi 28 560 2300~~

~~BWR 8X8 1977~ Hitachi 38 6900 38200~~

~~124 Table UT. 4 Typical Fuel Loading Scheme (4 loop PWR)~~

•Fuel Type ••••••-•--• •17X17 •Cycle Length •••••••• •12months •Fuel 23SU Enrichment •3.4wt*

~~Cycle N-l N N+l~~

~~Region Enrichment Load Discharge Load Discharge Load Discharge~~

~~3.4wt%+Gd* 1 1~~

~~Nl"t O^ 3. 4wt% 1 6 16~~

~~3. 4wt%+Gd' 60 59 1 1~~

~~£j * I N 9 3. 4wt% 28 12 16 1 6~~

~~3.4wt%+Gd* 60 0 60 59 1 1 N-l 3.4wt* 28 0 28 1 2 1 6 1 6~~

~~3.4wt%+Gd* 60 0 60 59 N 3. 4wt% 28 0 28 1 2~~

~~3. 4wt%+Gd* 60 0 N+l 3.4wt% 28 0~~

~~Cycle Burnup 14200 14200 14200 (MWd/t) Critical Boron cone 1390 1 390 1 390 at BOC (ppm) Number of Fresh Gd 960 960 960 Fuel Rods~~

*A Fuel Assembly Contains 16 rods of 1.9 wt%U235-6wt%Gd0 2 3

~~125 Table ffl.5 Typical Fuel Loading Schem loo2 ( e p PWR)~~

~~•Fuel Type -••-••••••• •14X14 •Cycle Length ••-••••• •12raonths •Fuel 236U Enrichment •3.5wt%~~

~~Cycle N-l N N-f 1~~

~~Region Fjirichment Load Discharge Load Discharge Load Discharge~~

~~N-4 3. 5wt%+Gd* 1 1~~

~~3.5wt*+Gd* 12 12 N-3 3. 5wt3S 28 27 1 1~~

~~3.5wt%+Gd* 12 0 1 2 12~~

* £ i N>( 2 3.5wt* 28 0 28 27 1 1

~~. 5wt%+Gd3 * 12 0 1 2 0 12 1 2 N-l 3.5wt* 28 0 28 0 28 27~~

~~3. 5wt%+Gd* 1 2 0 1 2 0 N 3.5wtX 28 0 28 0~~

~~3.5wt%+Gd* 12 0 N+l 3.5wt% 28 0~~

~~Cycle Burnup 11100 11100 11100 (Hid/t) Critical Boron cone 1 45 1 1451 1451 at BOC (ppm) Numbe f Freso r d hG 1 44 1 44 1 44 Fuel Rods~~

*A Fuel Assembly Contains 12 rods of 2.0 wt%U235-6wt%Gd203

~~126 Table ffl.6 Typical Fuel Loading Schem loo3 ( e p PWR) •Fuel Type -•-••••---• •15X15 •Cycle Length •••••••- •12months •Fuel 236U Enrichment •3.4wtX~~

~~Cycle N-l N N+l~~

~~Region Enrichment Load Discharge Load Discharge Load Discharge~~

~~3.4wt%+Gd* 17 17 N 3 3.4vrtK 20 20~~

~~4wt*+Gd. 3 * 40 23 1 7 1 7 j £ l ' 1 N 2 3.4wt» 20 0 20 20~~

~~3.4wtX+Gd' 40 0 40 23 17 1 7 N-l 3.4wt% 20 0 20 0 20 20~~

~~3.4wt!W5d* 40 0 40 23 N 3.4wtX 20 0 20 0~~

~~3.4wt%+Gd* 40 0 N+l 3.4wt% 20 0~~

~~Cycle Burnup 1 2300 1 2300 1 2300 (Iffld/t) Critical Boron cone 1112 1112 1112 at BOC (ppm) Numbe f Freso r d hG 640 640 640 Fuel Rods~~

*A Fuel Assembly Contains 16 rods of 1.9 wt^U235-6wt%Gd203

~~127 TABLE III.7. EXAMPL COMMERCIAF EO GADOLINIF O E LUS A FUEL ASSEMBLY~~

~~MIHAMA Uni1 t OHI Uni1 t OHI Uni2 t Cycl7 e Cycle 8 Cycl8 e~~

~~f Loopo . sNo 2 4 4~~

~~Fuel Assembly Total No. 121 193 193 Type 14X14 17X17 17X17~~

~~Criticality Date 1988.4.18 1988.6.15 1988.11.30~~

~~UÛ2 Assembly 28 48 72 No. Enrichment 3.2 3.4 3.4 (wt%)~~

~~Gadolinia Assembly 8 32 32 No.~~

~~Gadolinia Rod 12 16 16 A /F . No Enrichment 1.7 1.9 1.9 (wt%)~~

~~MOX Assembly 4 — — No.~~

~~PuU2 Enrichment 3.8 — — (Wt%)~~

~~128 TABLE m.8. STARTUP PHYSICS TEST RESULTS IN OPERATING PLANT WITH MANY GADOLINIA FUEL ASSEMBLIES~~

~~M1HAMA Unit 1 OH! Unit 1 OKI Uni2 t Test Item .Cycl e7 Cycle8 Cycle 8~~

~~Critical Boron AR4 O- AR8 O-1 ARO -11 Concentration Difference (ppm) 8 —1 n Di 5 Bi— n Din -14 (M-P)~~

~~Minimum Shutdown Margin:Boron 4 -10 19 Concentration Difference (ppm)~~

~~MTC Difference -1.3 -2.8 -0.1 (pcro/'C)~~

~~Rod Worth Ban —6.kB 85 1. Ban kD Bank D 1.1 Difference (%) Bank A -7.5 Bank C -7.1 Ban -3.kC 8 (M-P)/PX100~~

~~Boron Worth ARO ARO ARO Difference (%) i -7. l 5.43 1 -1.3 (M-P)/PX10n Di 0 Bin Din~~

~~Max. Error 6Q _38 Power Distribution /n\ *| f\\ D.U O.O 6.0 Difference 1/4 Core Power -, - (HZP, ARO) n nc 1.0 Tilt •~~

~~129 Table Et. 9 Summary of test fuel rod specifications~~

Test No. 510 511 . 512 Fuel Type U fuel Gd Fuel U fuel Item 17X17 PWR Type TypR PH e14X14 Diameter (mm) 8.19 9.29 Length (mm) 13.5 10.0 U Enrichmen} (% t 20.0 20.0 / i c . i x c r End Shape dished unchanfered chamfered SGd 0 ~6 0

Gd160 Enrichment (w/o) ~0fl C Material Zircaloy-4 Cladding Outer Diameter (ram) 9.50 10.72 Thickness (mm) 0.572 0.62 Fuel Stack Length (mm) ~80 Fuel Rod s FilGa l 1 atm.He Gap ïïidth (mm) 0.083 0.095

~~Table ffl.10 Summar f teso y t resits~~

~~Failure or No Failure Energy Failure Test No. Fuel Type No Post-test Fuel Rod Appearance Deposition Frag- Failure Chack Fracture (cal/g.UOï) mentation~~

~~510 - 1 17X17 252 O Uniform oxidation -3 PWR Type 264 O Circumferential cracks~~

~~— 2 U02 276 O Circumferential cracks~~

511 -1 232 O Uniform oxidation -3 265 O Oxidation, Slight bending -4 17X17 275 O Circumferential cracks -2 PTO Type 293 O Fracture into three pieces -7 GdzOa-UOz 351 O Fracture into many pieces -6 386 O Fragmentation -5 443 O Fragmentation

~~512 -2 14X14 217 O Uniform oxidation -3 P»R Type 232 O Uniform oxidation~~

~~- 1 U02 246 O Circumferential cracks~~

~~130 Table IÏÏ.11 Fuel rod failure thresholds~~

Test Fuel Rod Type Threshold Energy Series TypO PT e 7 1X1 7 510 252 —264 cal/g.U02 2 fued U0 ro l 17 x 17 PWR Type 511 265 —275 cal/g.U02 Gd203-U02 fued ro l 14 X14 PWR Type 512 232 —246 cal/g.U02 U02 fuel rod

131 Plenum Spring Claddirig Tube N) Upper End plug ' Lowed En r Plug Fued lRo a-— (f1— ID —— — " ŒP - -— ._ . ——————i,7,,v,,l— ii--.. •• it'-. i ii i • Pellet -.,,., .... Fffpr.tivfi —————————————————— >

~~Channel Fastener Expansion Spring Lower Tie Plate~~

~~Flow Hole~~

~~W : Water Rod T : Tie Rod~~

Flg. III.1. Latest BWR fuel design high( burnup 8x8) oooooooo © © © © © oooooooo oooooooo ooo ooo ooo oooooooo oooo oooo ©©©©©@( oooooooo c oooooooo oooooooo oooooooo oooviooo oooooooo OOOOOOOO vvQOOQQQQQyy

~~Highest Enrichment Lowest Enrichment Enrichment decreases as number increases~~

~~Fig. 111.2. Typical rod array of BWRfuel (highbumup 8x8)~~

~~133 u 4.S 05~~

~~Lower Nozzle Fuel Rod -Js Upper Nozzle Fuel Rod Hold down Spring RCCA Guide Thimble~~

~~M.D. Guide Thimble~~

~~C-C B-B A-A~~

~~17X17 A-Type Fuel~~

~~&•• 4.0 S~~

~~/ •Fuel Rod Nozzle Hold down Spring Fued RCCRo l A Guide Thimble~~

~~-H.D. Guide Thimble~~

~~C-C B-B A-A~~

~~17X17 B-Type Fuel~~

~~Fig. 111.3. 77x 77 type PWR fuels~~

~~134 HGFEDCBA~~

~~2 - N - - - N-1 > ••( N-1 V - N-l)---(N-) <§ - l>-~~

~~N-2 N~~

~~10 N- l N-2 N-1 N~~

~~11 N~~

~~12 N-2 N-2 N-1~~

~~13 N N-l~~

~~\ 14 N-1 N-1 N-1~~

~~15 N N N~~

Mark Region Number Numbe f Asso r 'y (S3) 12A(3. 4wt%+Gd*) 1 (Twice Burned) N-2 12B(3.4wt%) 16 (Twice Burned) (ÊM) 13A(3. 4wt%+Gd*} 60 (Once Burned) N-1 13B(3.4wtX) 28 (Once Burned) N 14A(3. 4wt%+Gd*) 60 (Fresh Fuel) N 14B(3. 4wt%) 28 (Fresh Fuel)

*A Fuel Assembly contains 16 FRs of 1.9w/o U235 - 6w/o Gd203

~~Fig. 111.4. Fuel loading pattern Equilibrium cycle of 4 loop Gd fuel core~~

~~135 Critical Boron Cone.~~

~~1500~~

~~V m pp 0 9 x l3~~

~~1000~~

~~500 -~~

~~5000 10000 15000~~

~~Burnup (MWd/t) Fig. 111.5. Critical boron concentration burnup. vs Equilibrium cycle loop4 fuelf d o G core HFP, AfiO~~

~~HGFEDCBA~~

~~0.84 0.97 1.13 1.20 1.20 1.1 7 1.15 0.73~~

~~0.98 1.10 1.18 1.20 1.18 1.2 0 1.13 0.95 =»= 10 =4= 1.19 1.22 1.16 1.1 6 1.1 1 1.1 1 0.92 11 1.20 1.20 1.16 1.09 1.04 0.96 0.84 0.62 ' — 1 12 1.20 1.18 LIS 1.04 0.91 0.8 1 0:60 1~~

~~1.17 1.20 Ll'O 0.9 S 0.86 0.9 1 0.42 1 -••• 14 1.15 1.12 1.1 1 0.83 0.60 0.4 1 =4= 15 0.73 0.95 0.92 0.6 1~~

~~Fig. IH.6. Radial power distribution loops4 f fueld o G core BOG, HFP,ARO~~

~~136 1.5 r ,1.48 (Max)~~

~~1.4~~

~~1.3~~

~~1.2~~

~~1.1~~

~~1.0 5000 10000 • 15000~~

~~Burnup (MWd/t)~~

~~Fig. 111.7. Fxyvs.burnup 4 loop, equilibrium cycle, Gd fuel core~~

~~137 7 10 11 12 13 tx t~~

~~3 ---N-N- l ---N-2 --(N-2>-- ... N_ 2 --- N~~

~~H N-l N-2 N- 2 N- 1 N~~

~~I N-2 N-2 N-2 N-l N~~

~~N- 2 N-l N~~

~~K N-l N- 1~~

~~L N- 2 N- l N N~~

~~M N N~~

Mark Region Number Number of Ass 'y N-3 . 5wt%1(3 6 ) 3 Cycl( 1 e Burned) ce> 178(3. 5wt%+Gd*) 12 (Twice Burned) N-2 17A(3.5wt%) 28 (Twice Burned) CED- 18B(3.5wt%iGd*) 12 (Once Burned) N-1 ISA (3. 5wt%) 28 (Once Burned) N 19A(3. 5wt%iGd*) 12 (Fresh Fuel) N 19B(3. 5wt%) 28 (Fresh Fuel)

~~Fuel Assembl Ow/. 2 yo f o containU23 s - 6w/5FR o2 1 s Gd203~~

~~Fig. 111.8. Fuel loading pattern Equilibrium cycle loop2 fuelof Gd core~~

~~138 1500~~

~~1000~~

~~o o o c o o CQ~~

~~CO E U O, •rH Q, o~~

~~5000 10000 15000 Burnup (MWd/t)~~

~~Fig. 111.9. Critical boron concentration vs. burnup Equilibrium cycle loop2 fuelof Gd core HPF,ARO~~

~~139 7 3 1 2 1 1 1 10 V t G~~

~~H 1.17 1.01 1.08 1.13 0.99 1.08 0.67~~

~~1.12 1.08 1.18 1.10 1.26 1.06~~

~~0.96 1.13 1.10 1.17 x 0.96 0.70~~

~~K 1.04 0.99 1.26 0.96 0.59 \ L 0.89 1.08 1.06 0.70~~

~~M 0.80 0.67~~

~~= 1.38 in (K-9)~~

~~Fig. 111.10. Radial power distribution loop2 fuel f d o G core BOC, HFP,ARO~~

~~1.5 1.48 (Max)~~

~~1.4~~

~~1.3~~

~~1.1~~

~~1.0 5000 10000 15000~~

~~Burnup (MWd/t)~~

~~Fig. . burnup111.11vs y .Fx loop,2 equilibrium cycle, fuelGd core~~

~~140 ^ H G F D C B A 8t-~~

~~10 N-2~~

~~11 N-2~~

~~13 N-2 N-1~~

~~14 N- l N-1 N~~

~~15 N N~~

Mark Region Number y ' s NumbeAs f o r (ED 18A(3.4wt%+Gd*) 17 (Twice Burned) N-2 1 8B (3. 4wt%) 20 (Twice Burned) (Fl) 19A(3.4wt%+Gd*)' 40 (Once Burned) N- 14wtX. (3 B )19 20 (Once Burned) N 20A(3.4wt%+Gd*) 40 (Fresh Fuel) N 20B(3.4wt%) 20 (Fresh Fuel)

~~A Fue* l Assembly f 1.9w/o contain s FR o 6 Ü231 s - 6w/5 o Gd203~~

~~Fig. 111.12. Fuel loading pattern Equilibrium cycle of 3 loop Gd fuel core~~

~~141 5000 10000~~

~~Burnup (MWd/t)~~

~~Fig. 111.13. Critical boron concentration burnup. vs Equilibrium cycle of 3 loop Gd fuel core HPF, ARO~~

~~HGFEDCBA~~

~~. 1.13 =t=±~~

~~1.29 1.29 1 \~~

~~10 1.13 1.22 1.10 1~~

~~11 1.19 1.13 1.21 1.16 \, 1~~

~~12 0.98 1.08 0.94 0.97 0.83~~

~~1 \~~

~~13 1.18 1.12 0.97 0.83 0.55 =t=~~

~~14 1.26 1.02 0.82 0.61 1 15 1.04 0.78~~

~~Fx = 1.4y n i 1(G-9 ) Fig. III.14. Radial power distribution loop 3 fuel d f o G core BOG, HFP, ARO 142 1.5 r- 1.48 (Max)~~

~~1.4 v Fxy~~

~~1.3~~

~~1.2~~

~~1.1~~

~~1.0 5000 10000 15000~~

~~Burnup (MWd/t)~~

~~Fig. 111.15. Fxy vs. burnup loop,3 equilibrium cycle, fuelGd core~~

~~Hot full power. «II rods out —— Calculated • Measured~~

~~5000 10000 15000 Cycle burnup (MW-d/t)~~

~~Fig. 111.16. Critical boron concentration~~

~~143 1.0~~

~~»O) o Q- .-S' Prediction "i (Bank D-215 step) O> ,8-' fi-fi- OHI2 Cy5, GdAssy, C-03. N-13 0.6 (O) (A)~~

~~i__i__i \__i__t__t__i__i i t__i J 0.0 5000 10000 15000 Cycle Burnup (MwdA)~~

~~Fig. 111.17. Gd FA power vs. cycle burnup~~

~~XIO~~

~~270-l ta 4-» 2 240- 210- ' "" *• ""."• * • " •*""*•" '~ I 180 ISO- 120 (CPS) 90J~~

~~60 \ 30-~~

~~0 80 120 0 40 1600 0 2000 2400 2800 3200 3600 top , \ bottom distanc) e(m frop mto~~

~~Fig. 111.18. Axial gamma scanningR F d G f o~~

~~CCTS)~~

~~500 -EE-. :a P 7Ë3=:: =~~

~~315*«- 135* T 0 Lr. LKr—lO-LJ J. Li-J.-J-t I J i~~

~~Fig. 111.19. M/cro gamma scanning(grossR F d G ) f o~~

145 iUU o : U fuel pellet 99 fued G l • pelle * t 98 Q 97 -£*»«• £ 96 -fr tO 95 IX" ^"T^--*.. U pellet irradiation swelling rate "0 93 *— T „ _ f\ t)O/ ( A \7 1 \T\ / ( 1 rt20// /^nr^ I I Q. / l J v>ll / \ X iv / / T / r i / \ \J»£t ^ 1^ Q / 92 - u_ 91

~~90 i i i l i l i 1 i 3 10 20 30 40 5 Fuel pellet burnup (GWd/t)~~

~~Fig. 111.20. Pellet density change with burnup~~

~~U Gd pellet pellet T Pellet cente • r v o 15 - Pellet E peripheral~~

~~C 10~~

~~Q • 0 ö ° 0 - * <~~

~~5 ~° 0 T T~~

~~Averag e gra i v ™~~

~~± L 0 ' L 0 4 0 3 0 2 0 1 0 50 Burnup (GWd/t)~~

~~Fig. 111.21. Change of pellet average grain size with burnup~~

~~146 30 T 1 JJJ „. rt l F3A1 & F3A6 v UO2-Gd2O3 J assemblies p} 3A 7j assemblieo 2 *IS s n o UOr-GdzOsJ 20 v *!*?,.,« 1F-3A3 & F3A9 •p oU02-Gd2031 assemblies CO~~

~~TT O 0 40 0 30 0 20 0 10 Maximum linear heat rate (W/cm) (maximum experienced power at>10GWd/t)~~

~~Fig. 111.22. Dependency of fuel rod FGR rate on LHGR~~

~~v Big Rock Point 40 A Dresden 3 O Oyster Creek D Peach Bottom 2 30 0 CD test data -P CC Q 20 - ° $x> 0 O ° Q 0 O~~

° a v 10 ° Oa o w v o V o> ft ° ^ 0 \7———— —OXIÖÖ* — —1 — <3^<5& C?— I— —x&^o-o— — i ————— 0 5 7 10 15nrr>7n 3 20 0 3 25 Bumup (GWd/t) Fig. 111.23. Fuel rod FGR rate vs. burnup

~~Pelle2 UÛ t % 32 s rin Top end fitting Pelle2 U0 Bottot fittin% d 20%.UQ32 men g 2 Pelle— Ptc— 9/ Qodding~~

~~•(POOL~~

~~-Active length (81mm)~~

~~———————— Overall length (229mm)- Fig. IIJ.24. Schematic view of test FR~~

~~147 Offset loading tube û~~

~~+OOOO- + + - -t H + •*••« •»+ -t • + + + +OOOOOOOO+ + -n-~~

~~Transient rod (3(\ V/°o?O^g^O®c|_~~

In^men \ V^^^^^W^ *

~~Regulatin (6)d ro g' f + + + + + Safety rod (2)~~

~~Fig. 111.25. Standard operating core configuration~~

t-v^WUV 1 1 1 1 1 I^U 5 20000 NVT loS o A / i ^ y / 80 - ^ 15000 I A }— NV / M ^ fc_ 1 1 \ 40 ö ff il \ cc <^ ii \ | 5000 II \ :>> 1 / \ 20 fc a> \ / / i5 a. n ^/ J \_ \J ,.,...... ,.. o I o i i i i * 0 63 0 62 0 61 0 60 0 59 0 58 0 57 Time (msec)

~~Fig. III.26. Pulse reactor power and core energy release for 4.67 $ pulse~~

~~148 Test ) Series \ Tvod FueRo el Threshold Energy 17X17 PWR Type 510 252 -264 cal/g.UO, UO2 fuel rod i 17x17 PWR Type 511 265-275 cal/g.UOa Gd203-U02 fuel rod~~

~~14X14 PWR Type ,12 232-246 cal/g.UO, 2 d fuero U0 l~~

~~Oxide flake~~

~~bot torn top~~

~~20mm~~

~~(a) Test Na 51 0-1 ( 252 cal/g.UO? )~~

~~top bottom~~

~~\ :?ractured during fuel telulling~~

~~(b) Test No. 510-2 ( 276 cal/g.U02 ) Fig. 111.27. Representative post-test FR appearances in No.510 test series, showing oxide (a) melt (b) flake projectionand~~

149 Max. Average • O 14x14 Type 10%E U0 Fuel rod 10 2 q * 017x1 2 Fued presenro U0 l 7 Typt e « GdaOj-UO / / wor ?> A Fued k ro lA r 3 A A // « GdgOj-UOj Riel rod/1' a 14x1 • 4 Type UOz Fuel rodj - 8 £: 7

~~c 6 o 1 5~~

~~«*§4— h -. * OJ Q 3 o r e A 2 2 -§ - -* * • ^ o or I 0 i t i _I 1 I 0 30 0 25 0 20 150~~

~~Energy Deposition (cal /g-U02)~~

~~Fig. 111.28. Comparison f radialo deformation~~

~~150 Appendix IV GADOLINIA FUEL UTILIZATION IN THE REPUBLIC OF KOREA~~

~~This Appendix provides additional details on the experience with Gd in Korea, as summarize Chapten di . 5 r~~

~~utilizatiod V.lI G . n~~

~~KAERI has contracts with KWU for reload core fuel design for all operating reactors and with ABB/CE for initial core fuel design for the Yonggwang 3 and 4 units.~~

Table IV.1 compares various nuclear parameter reloan si d core designn a s A . exampl reloaa f eo d core design Kori-n ,i 2 cycl e7 refuellin g (low-leakage loading pattern), 52 conventional standarreplacee ar 4 fress y FA b dh standar - s (3.4U FA do 1w/ enrichment) and 48 fresh KOFAs (designed by KAERI and KWU). Out of these 48 KOFAs, 28 fuel assemblies contain Gd FRs of 1.8 - 6 w/o Gd. The length of cycle 7 is 14,420 MWd/tU (375 EFPD), compared wit cycle hth lengte6 f 13,10ho 0 MWd/tU (348 EFPD).

Table IV.2 shows the FA configurations that contain Gd FRs. The design specifications require that Gd FRs not be located directly adjacent to water holes. Table IV.3 shows the FA design parameters. Figures IV.1 through IV. 16 present the various burnable absorber patterns in the reload cores.

In ChapteFigurn i infinit e 8 th e 4. , 4 re multiplication YGN-3/e factoth r fo r 4 initial core is shown for the case of an 8 shimmed fuel assembly. Figure IV. 17 shows the same for the case of a 12-shimmed fuel assembly. Figure IV.18 and Figure IV.19 compare the Gd fractions remainin shimme2 1 functioa d s ga an d8 f burnufueno e th l r assembliespfo , respectively.

Table IV.4 shows the number of Gd FRs and the number of FAs with Gd FRs, either already loade reactordn i deliverer so plante f Octobeth o s do st a r 1993. Table IV.5 details numbe e brokes th d FA f an n o r s dow numbee FR th d n G int f numbee o ro th f residenco r e e cyclevariouth t a s loades s o Octobet FR plants p e numbeu d Th G r . f o 199r 3 corresponds to 26 reactor cycles.

~~alreads hav372e s FR th d FR eyf 8 G o d loadedbee t G nOu 4 discharge86 , d from storew poole th no show s n sde a i ar core Tabln d i th ean s e IV.6FR d G . Some th f eo completed the third cycle.~~

Poolsid showes e inspectionFR noticeabld o dn G e th f so e anomalies afte thire th rd cycle. These poolside inspection performee sar d usin equipmene gth t develope KAERIy db , called "Versatile Fuel Examination Stand" (VFES) poolside Th . e examination with VFES focuse n measurino s e differencgth n dimensioni e s between as-fabricated datd an a irradiated data, and on verifying the performance of the fuel during irradiation.

~~IV.2 Fuel Modelling~~

~~Initial core The fuel modelling for initial cores of YGN 3 and 4 is described in Chapter 4. Figures IV.20 through IV.23 sho code wth e system initia e useth dn li core desig YGNf no - 3/4.~~

~~151 Reload core~~

~~Five main computer codes are used in Korea by KAERI and KWU in nuclear reload fuel design, includin FRsd gG . Figure IV.24 show calculationae sth l flow diagra r fuemfo l assembly design.~~

grou5 8 , pD FASE cell-burnu1 a Rs i p programm latticr efo e configurations basen do a heterogeneous version of the slowing down programme MUFT and a modified version of the thermalization code THERMOS. FASER generates basic nuclear cross section librarie r reactofo s r simulation s functiona s f o nuclides , burnup, soluble boron concentration, coolant density/temperature and fuel rod temperature.

MULTIMEDIUM is a 2-dimensional assembly code for calculation of pinwise flux, power and burnup distributions for PWR fuel assemblies. The maximum number of energy flue xTh groupdistribution . 10 s determinei e b n sca d eithe usiny b rnoda e gth l expansion method (NEM r diffusiosolvinD )fo 2 e gth n theory equations applyiny b r o , nodae gth l discrete ordinate method (NDOMe approximateth r fo ) transpordD 2 solutio e th f tno equations with isotropic scattering maie Th n. programme usagth f eo generato t s ei e heterogeneous fuel assembly form functions and heterogeneity factors for the second homogenization process.

thred MEDIUM-an e o dimensionatw a s 2i l reactor burnu fued pan l management code for the prediction of long-term reactivity behaviour, global power distributions, control rod worths, critical boron concentrations, reactivity coefficient d othean s r design parameters multi-grouD 3 r o t solveD I 2 . p e diffusiosth n theory equation Cartesian si n geometry by applying a high order coarse-mesh nodal expansion method (NEM) and nonlinear feedback correction (NEMFC second-ordee th y b ) r representatio f macroscopino c cross-section variation nodea n si . MEDIUM-2 also calculate node-averagee sth d values fluxe sucth ,s ha power , iodin xenod ean n quantities.

PINPO assembln a Ws i y dehomogenization cod reconstructior efo f locano l flud xan FR power distribution from the nodal coarse-mesh solution by MEDIUM-2. The programme determine locae sth l continuous flu reactoplane xY th high-ordesolutio f a X- e o y e rb th nn i r non-separable interpolation method. PINPOW takes into account the node burnup shapes, whic e providear h y MEDIUM-2b d o determint , e local burnup distributionse Th . heterogeneous fuel assembly form functions provided by MULTIMEDIUM are used to determine the pinwise flux, power and burnup distributions by using a modulation technique. PINPOW can determine the reaction rates of the incore measurement system and generates all theoretical information needed by the process computer to determine power and burnup distributions.

PANBOX is the reactor dynamics program system used for reload safety analysis an l kinddal transientf so whicsn i powee hth r distributio significantlns i y affected. These transients include short term accidents ejectiolikd ero lond ngan term events like xenon redistribution.

~~PANBOe Th X program system consist: sof~~

~~DIEB: - input processor for IQSBOX~~

~~data base generator for APCBOX~~

~~IQSBOX - LWR core transients~~

~~APCBOX - LWR xenon dynamics~~

152 IQSBOX solve two-groue sth p space-time-dependent neutron diffusion equatioy nb the nodal expansion method (NEM) or nodal integration method (NIM) in 3D Cartesian co- ordinates, coupled wit efficienn ha t time integration procedure prograe Th . mespecialls i y suited for the calculation of steady state and transient power distributions under off-normal core conditions. APCBOX is the 1D version of IQSBOX used for xenon dynamics calculations.

~~153 Ul TABLE IV.1. COMPARISON OF NUCLEAR PARAMETERS IN RELOAD CORE DESIGN~~

~~Kor1 - i 4 Kor& 3 i- Yonggwan2 & 1 g - ITEMS W(OFA) KOFA* W(OFA) KOFA W(OFA) KOFA~~

~~Fuel enrichmen) o / w ( t 3.2 3.5 3.2 3.7 3.2 3.7~~

Region average disch, 35500 35000 37000 42000 37000 42000 Bu(MWd/tU) Maximum rod average, 47000 52000 52000 Bu, (MWd/tU) Fq 2.08 2.08 2.30 2.30 2.30 2.32 Limit Fxy 1.55 1.55 1.55 1.55 1.55 1.55 uncertai% +8 n

~~Delayed neutron 0.0072/0.0043 0.0071/0.0049 0.0075/0.0044 0.0068/0.0049 0.0075/0.0044 0.0068/0.0049 Limit fraction (BOG /HOC)~~

~~Neutron Lifetime 26 28 18.1 /19.4 17/28 21.3/21.7 17/28 Limit (A/sec) (BOC/EOC)~~

~~) AF(% D -5 —- +5 -5 + — » -12—» +3 -12—» +3 -13 2 + — - -12 —- +3 Limit~~

~~BA material B4C (13.5) Gd2O3 (8/1.8) B4C (13.5) Gd2O3 (8/1.8) B4C (13.5) Gd203 (8/1.8) (Gd, %/U-235 w/o)~~

~~DTC (pcm/'C) -5.2—- -1.6 -5.6 —- -2.5 -5.2—» -1.6 -6.0—- -2.8 -5.2 —— -1.6 -6.0—- -2.8 Limit~~

~~MTC (pcm/°C) +9.0 — - -59.4 + 10.0—» -64.1 +9.0—» -72.0 +10.0—» -71.3 < +9.0 +10.0 —- -71.6 Limit~~

Doppler-onlC yP U: -19.4/ -12.6 U: -16.4 / -14.0 U:-19.4/-12.6 U: -14.3 -12./ 5 U:-19.4/-12.6 U:-14.3/-12.5 Limit ( pcm / % power ) L: -9.6 / -6.1 L-11.2/-9.0 L: -9.6 -6./ 1 L: -9.8 -8./ 2 L: -9.5 -6./ 0 L: -9.8 / -8.2

~~Boron coefficient Limit ( pcm / ppm ) -16 —- -7 -146 - — » -16 — - -7 -146 - — » -17 —--7 -13 6 —- - Contro< l Rods> EOC Total-stuck rod worth ( % ) 5.75 5.99 9.01 7.55 8.37 7.48~~

~~Control Reactivit) (% y 3.09 3.35 3.26 3.55 3.49 3.58~~

~~SD) M(% 2.09 2.04 4.85 3.24 4.04 3.15 KOFA: Korean Fuel Assembly TABLE IV.2. Gd FR DESIGN DATA~~

~~14x14 16x16 17x17 type type type~~

~~Active) lengtm m ( h 3658 3658 3658~~

~~Rod) lengtm m ( h 3852 3847 3847~~

~~Cladding O.D. (mm) 10.75 9.5 9.5~~

~~Rod pitch ( mm ) 14.12 12.32 12.6~~

~~Assembly width ( mm ) 1972 1972 2140~~

~~Gd203) conteno / w ( t 6.0 6.0 9.0~~

~~U235enrichment (w/o) 1.8 1.8 1.8~~

~~3 UO2- Gd2O3 density ( g / cm ) 10.08 10.08 10.08~~

~~Table IV.3. Fuel assembly design parameters~~

~~Kori 2 Cycle 7~~

~~Region 6s 7A-S 7B 8A 8A-S 8B 9 9G~~

~~Enrichment (w/o U235) 3.40 3.40 3.61 3.41 3.41 3.62 3.50 3.47~~

~~Number of Assemblies 4 4 13 32 4 16 28 20~~

~~Approximate Burnup 17.9 14.9 25.5 15.1 0.0 15.0 0.0 0.0 at BOC 7 (GWd/tU)~~

~~All regions contain as-built data except region 9 and 9G.~~

~~Burnable Absorber Rods~~

~~Cycle 6 Cycle 7~~

~~Absorber Material Pyrex (BaOa) fued G l~~

~~Content, w/o 12.5 6.0~~

~~Numbe Corn i r e 176 80~~

~~155 Tabls FR Numbeed wits d an IV.4h G FA s f o .rFR Numbe d G f o r Octobeo t p u r 1993~~

~~Delivered Loaded t loadeye t dno t Bu~~

~~s FR d G s FA d G s FR d G Gd FAs~~

~~3728 640 784 104~~

~~Table IV.5. Number of Residence Cycles of Gd FRs Currently in Operation (October 1993)~~

~~Plant Kori Kori YongGwang Ulchin Unit 1 2 3 4 1 2 1 2 Total~~

~~Cycle length(month) 18 18 12 12 12 12 12 12~~

~~Gd Rods 64 0 336 384 256 288 112 112 1552 1 FAs 16 0 52 48 36 40 20 20 232~~

~~# of Gd Rods 64 0 128 288 112 112 112 112 928 residence 2 cycles FAs 16 0 24 36 20 20 20 20 156 Gd Rods 84 84 104 112 384 3 FAs 13 13 21 20 67~~

~~Table IV.6 Stores .FR Numbe Poon d i G f l o r(Octobe r 1993)~~

~~Plant Kori Kori YongGwang Ulchin Total Unit 1 2 3 4 1 2 1 2~~

~~Gd Rods 112 112 1 FAs 28 28~~

~~# of Gd Rods 64 192 28 188 28 24 524 residence 2 cycles FAs 16 48 7 35 7 3 116 Gd Rods 4 112 112 228 3 FAs 1 20 20 41~~

~~156 270°~~

~~4 4 C~~

~~4 4 4 D~~

~~4 4 E~~

~~4 4 F~~

~~180° 4 4 G~~

~~4 4 H~~

~~4 4 I~~

~~4 4 4 J~~

~~4 4 K~~

~~M 90'~~

~~3 1 2 1 1 1 0 1 9 8 7 6 5 4 3 2 1~~

~~Numbers Indicate Number of Gd FRs Total : 80 Gd FRs 6.0w/o Gd2O3/ 1.80w/o U235~~

~~Figure IV.1. Burnable Absorber Pattern (Region 9G):Kori-2 Cycle7~~

~~contro guidd ro l e tube~~

~~instrumentation tube~~

~~FR with Gd2O3~~

~~Figure IV.2. Burnable Absorber Configuration Within FA : Kori-2 Cycle 7~~

~~157 270°~~

~~4 4 C~~

~~4 4 D~~

~~4 4 E~~

~~4 4 F~~

~~180° G~~

~~4 4 H~~

~~4 4 1~~

~~4 4 J~~

~~4 4 K~~

~~90e~~

~~1 2 3 4 5 6 7 8 9 10 11 12 13~~

~~Burnable Absorber Pattern Numbers Indicate Number of Gd FRs Total : 64 Gd FRs 6.0w/o Gd2O3/ 1.80w/o U235~~

~~Figure IV.3. Burnable Absorber Pattern (Region 13G):Kori-1 Cycle 11~~

~~guide tube~~

~~instrumentation tube~~

~~Gd FR~~

~~Figure IV.4. Burnable Absorber Configuration Within FA : Kori-1 Cycle 11~~

~~158 RPNMLKJHGFEDCBA~~

~~180° 1~~

~~4 2~~

~~8 8 3~~

~~4 4 4~~

~~8 8 6~~

~~4 4 7~~

~~90' 4 4 8~~

~~4 4 9~~

~~8 8 10~~

~~4 4 12~~

~~8 8 13~~

~~4 14~~

~~0° Burnable Absorber Pattern Numbers Indicats e NumbeFR d G f o r Total : 112 Gd-FRs 9.0w/o Gd2O3 / 1.8w/o U235~~

~~Figure IV.5. Burnable Absorber Pattern (Region 8G-4,8G-8):Kori-3 Cycle6~~

~~159 D FR~~

~~guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~(4 Gd Assembly)~~

~~D FR~~

~~Q guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~Assemblyd (8G )~~

~~Figure IV.6. Burnable Absorber Configurations Within FA : Kori-3 Cycle 6~~

~~160 RPNMLKJHGFEDCBA~~

~~180° 1~~

~~8 2~~

~~4 4 4~~

~~4 4 7~~

~~90 ' 8 8 8~~

~~4 4 9~~

~~4 4 12~~

~~8 14~~

~~0° Burnable Absorber Pattern Numbers Indicate Number of Gd FRs~~

~~FRd G s Tota 4 9.0w/6 : l o Gd2O3/ 1.80w/o U235~~

~~Figure IV.7. Burnable Absorber Pattern (Regions 7G-4,7G-8):Kori-4 Cycle5~~

~~161 D~~

~~guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~Assemblyd G 4 ( )~~

~~D FR~~

~~Q guide tube~~

~~D instrumentation tube~~

~~• Gd FR~~

~~Assemblyd G 8 ( )~~

~~Figure IV.8. Burnable Absorber Configurations Withi : Kori- A nF 4 Cycle5~~

~~162 RPNMLKJHGFEDCBA~~

~~180° 1~~

~~4 2~~

~~8 8 3~~

~~4 4 4~~

~~8 8 6~~

~~4 4 7~~

~~90 ' 4 4 8~~

~~4 4 9~~

~~8 8 10~~

~~4 4 12~~

~~8 8 13~~

~~4 14~~

~~Burnable Absorber Pattern Numbers Indicats e NumbeFR d G f o r~~

~~9.0w/s FR d o G Gdp 2 Tota11 3: l/ 1.80w/o U235~~

~~Figure IV.9. Burnable Absorber Pattern (Regions 7G-4,7G-8):Yonggwang-1 Cycle5~~

~~163 FR~~

~~guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~Assemblyd G 4 ( )~~

guide tube D **

~~instrumentation tube~~

~~Gd FR~~

~~Assemblyd (8G )~~

~~Figure IV.10. Burnable Absorber Configurations Within FA : Yonggwang-1 Cycle 5~~

~~164 RPNMLKJHGFEDCBA~~

~~18Cf 1~~

~~4 2~~

~~8 8 3~~

~~4 4 4~~

~~8 8 6~~

~~4 4 7~~

~~90C 4 4 8~~

~~4 4 9~~

~~8 8 10~~

~~4 4 12~~

~~8 8 13~~

~~4 14~~

~~0° Burnable Absorber Pattern Numbers Indicats e NumbeFR d G f o r~~

~~9.0w/s FR d o G Gdp2 Tota11 : 3l / 1.80w/o U235~~

~~Figure IV.11. Burnable Absorber Pattern (Regions 6G-4,6G-8):Yonggwang-2 Cycle 4~~

~~165 FR~~

~~guide tube~~

~~D instrumentation tube~~

~~R F d G~~

~~Assemblyd G 4 ( )~~

~~guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~(8 Gd Assembly)~~

~~Figure IV.12. Burnable Absorber Configurations Within FA : Yonggwang-2 Cycle 4~~

~~166 RPNMLKJHGFEDCBA~~

~~180° 1~~

~~4 2~~

~~4 4 3~~

~~8 8 4~~

~~4 4 6~~

~~8 8 7~~

~~90' 4 4 8~~

~~8 8 g~~

~~4 4 10 11~~

~~8 8 12~~

~~4 4 13~~

~~4 14~~

~~0° Burnable Absorber Pattern Numbers Indicate Number of Gd FRs~~

~~Total : 112 Gd FRs 9.0w/o Gdp3/ 1.80w/o U235~~

~~Figure IV.13. Burnable Absorber Pattern (Regions 5G-4,5G-8):Ulchin-1 Cycle 3~~

~~167 D FR~~

~~guide tube~~

~~instrumentation tube~~

~~R F d G~~

~~Assemblyd G 4 ( )~~

~~R F D~~

~~Q guide tube~~

~~Q instrumentation tube~~

~~• Gd FR~~

~~(8 Gd Assembly)~~

~~Figure IV.14. Burnable Absorber Configurations Within FA : Ulchin-1 Cycle 3~~

~~168 RPNMLKJHGFEDCBA~~

~~180° 1~~

~~4 4 4 2~~

~~8 8 3~~

~~8 8 6~~

~~4 4 7~~

~~90' 4 4 8~~

~~4 4 9~~

~~8 8 10~~

~~8 8 13~~

~~4 4 4 14~~

~~0° Burnable Absorber Pattern Numbers Indicats e NumbeFR d G f o r Total : 112 Gd FRs 9.0w/o Gd2O3/ 1.80w/o U235~~

~~Figure IV.15. Burnable Absorber Pattern (Regions 4G-4,4G-8):Ulchin-2 Cycle 2~~

~~169 D FR~~

~~[~] guide tube~~

~~instrumentation tube~~

~~Gd FR~~

~~Assemblyd G 4 ( )~~

~~guide tube~~

~~instrumentation tube~~

~~Gd FR~~

~~(8 Gd Assembly)~~

~~Figure IV.16. Burnable Absorber Configurations Within FA : Ulchin-2 Cycle 2~~

~~170 Effective Remaining Fraction Infinite Multiplication Factor lu c o-~~

~~o -4» 5'~~

~~7 Q 5' 8- o' a M o~~

~~O4 •^ 5* S- o Q_' o O O 03~~

~~O CL NJ aO w~~

~~n 2 Effective Remaining Fraction~~

~~(6 —. CL C~~

~~o Û. co —~~

~~15 Vf ro M-W ^ T*» c if //~~

~~—1 3 I/ c Ty -< TB» O 1 "0^ / O4 " > cz (1 -r 5' (1 •!• i-i{ 5 * Û* D K) 8-^ 1 S-V O O c\« 9>. — o [ 1 c hû to •f^ O CL ~n -f 1 M -p to co -i~~

~~3 r 1 DATA LIBRARY PREPARATION~~

~~PROCESSING CODES FLANGE DIT ETOG DATA LITHE LIBRARY GGCJ GROUP5 8 S RABBLE I DIT ASSEMBLY CODE~~

~~CROSS SECTION PROCESSING & TYPICAL SPECTRUM GEOMETRIES MIXING~~

~~z 'ASYMPTOTIC" FUEL CELL g SPECTRUM CALCULATIONS K FOR SELECTED SPECTRUM o in GEOMETRIES LU o F ü w SHIM AND FUEL CC CONDENSE CROSS SECTIONS (JJ < NEIGHBORS o Q IS TO FEW GROUPS~~

~~DEPLETIO SELECTEO NT D BURNUP OR EXPOSURE TIME~~

~~ASSEMBLY GEOMETRIES~~

~~EDIT DATA~~

~~1/4 ASSEMBLY PREPARE & SAVE DATA FILES~~

~~DATA FILTERR SFO SPATIAL DESIGN CODES~~

~~PERMITS RERUNNING PROBLE BURNUY AN T MA P POIN VARO TT Y SELECTED PARAMETER CONTINUR SO E DEPLETIOA N CHECKERBOARD WITH SYMMETRY~~

~~Figure IV.20. The DIT assembly code~~

~~173 CLAD~~

~~Figure IV.21. Typical DIT cell geometry for Gd fuel~~

~~DIT~~

~~FINE MESH CROSS SECTIONS CROSS SECTIONS~~

~~CURRENTS ROCS MC FLUXES~~

~~GLOBAL RESULTS INTRA ASSEMBLY RESULTS~~

~~Figure IV.22. FR peaking calculations in coarse-mesh~~

~~174 NOTES -~~

~~1 DONE ONLY AT THE START OF CALCULATION AND AFTER A DEPLETION~~

~~2 DONE ONLY WHEN REQUESTED~~

~~3 DEPLETABLE NUCLIDE AND FISSION PRODUCTS~~

~~Figure IV.23. ROCS logic flow~~

~~175 BADASA~~

~~N U KLAN~~

~~FASER Output XS-Data File Design Data (2nd File of the last run) FASAP LIBED~~

~~Tape 98 CA-XS P.o.C-XS MULTIMEDIUM~~

~~Het. form functions Tap3 e1 Het. factors Tape 10~~

~~WQFTT Tape 7 Tap3 e5 Data As Built i Tap0 e2~~

~~MEDIUM ITape 50 PINPOW~~

~~Figure IV.24. Calculation flow of fuel assembly design U use KAERy KW db d an I~~

~~176 Appendix V GADOLINIA FUEL UTILIZATION IN BELGIUM~~

~~This Appendix provides further details of the experience obtained in Belgium with Gd FAsreferres a , Chapten i o dt . 5 r~~

~~Tabl I summarizeeV. s maie sth FR nd characteristicG d an s FR U f bot so e hth irradiated in the BR3 reactor, as referred to in Section 5.1.1.~~

Figure V. 1 shows the total number of Gd FRs irradiated in the BR3 reactor between 1974 and 1987. Figure V.2 shows a typical evolution of power for two reactor cycles in BR3. Tabl 2 provideeV. s detailcommerciae th f o s Belgian i l irradiations FA n d G f o s reactors alss a , o describe Section di n 5.1.1.

~~177 TABLE V.1. MAIN CHARACTERISTIC IRRADIATEs PLAN3 FR BR d E TG TH D DN I AN U F SO~~

~~U FRs Gd FRs~~

~~U235 enrichment ( wt % ) 8.25~~

~~Gd2O3 content (wt%) 1.35-10~~

~~Pellet density ( % theoretical density TD ) 94-96~~

~~UO2 grain size (pm) 5-15~~

~~Gd2Os maximum particle size (p/m ) < 100~~

~~Pellet geometry Dished~~

~~ratiD L/ o 1.5~~

~~Oensification after res inte ring test < 1 % TD~~

~~Cladding~~

~~Material Zr4 stress-relieve fullr do y annealed~~

~~Outer diameter ( mm ) 9.5~~

~~Fuel Rod~~

~~Active fuel) lengtm m h( 1000~~

~~Diameter gap (jum) 200~~

~~Helium pressure ( kg / cm2 ) 11 -20~~

~~178 TABLE V.2. Gd UTILIZATION IN BELGIAN POWER PLANTS~~

~~Tihange 1: 900 MWe PWR 15x15 fuel~~

~~Cycle Loading Cycle Reload FAs Number Number % Gd Total length X of Gd of Gd Number (MWd/t) enrichment FAs FRs (21 of Gd FRs~~

~~7 1982 11945 8x3.20% 48x3.30% 4 8 8 32~~

~~8 1983 11619 8x3.30% 44x3.25% 16 8 8 128~~

~~9 1984 12510 52x3.25% 4 8 10 32~~

~~10 1985 16745 12x3.25% 12 8 8 384 60x3.45% 24 12 8~~

~~11 1986 14847 12x3.25% 24 8 8 192 56x3.45%~~

~~12 1988 12768 12x3.45% 8 8 8 64 40x3.90%~~

~~13 1989 15710 56x3.90% 20 8 8 160~~

~~14 1990 17075 60x3.90% 24 8 8 192~~

~~15 1991 13550 52x3.90% 16 8 8 128~~

~~16 1992 1 7400 (1) 63x3.90% 25 8 8 200~~

~~(1) Planned~~

~~(2) All schemes have Gd FRs at mid assembly positions.~~

~~179 TABLE V.2. (cont.)~~

~~Tihange 2: 900 MWe PWR 17x17 fuel Cycle Year Reload Cycle Numbef ro length Gd FRs (MWd/t)~~

~~5 1987 12390 128~~

~~6 1988 52 F As 13800 128 3.8% 16 Gd FAs~~

~~7 1989 s FA 2 5 13500 128 3.8% s FA d G 16~~

~~8 1990 52 FAs 12350 128 3.85% s FA d G 16~~

~~9 1991 52 FAs 15250 128 3.8% s FA d G 16~~

~~10 1992 s FA 2 5 14000 128 3.8% s FA d G 16~~

~~11 1993 60 FAs 14750 224 3.8% (estimate) 28 Gd FAs~~

~~positionind an A F r t mid-assemblga pe s FR FAs d d G G l I,n8 ai y sites~~

~~180 TABLE V.2. (cont.)~~

~~Tihange 3: 1050 MWe PWR 17x17 fuel~~

~~Cycle Year Reload Cycle length Numbef o r (including Gd FRs stretch) (MWd/t)~~

~~4 1988 52 FAs 12000 128 3.8% 16 Gd FAs~~

~~5 1989 52 FAs 14190 128 3.8% 1 6 Gd FAs~~

~~6 1990 52 FAs 13500 128 3.8% s FA d G 6 1~~

~~7 1991 52 FAs 14500 128 3.8% 1 6 Gd FAs~~

~~8 1992 s FA 2 5 13800 128 3.8% (estimate) 1 6 Gd FAs~~

~~In all Gd FAs, 8 Gd FRs per FA and positioning at mid-assembly sites~~

~~181 £81~~

~~Number of Gd rods ( % of core ) ro -^ o œ _j___i___i___i___i___, i~~

~~î CD o (A~~

~~T; (D~~

~~• £ • 5 S ; ''{.•- x c ••"•^ CD I UJ I CD l§ 3' O) clo aO D Î •>j CD~~

~~O. CD ^ S 500~~

~~400~~

~~§300~~

~~Œ 200 -~~

~~100 - Peak burnup: 42GWd/tM Peak burnup: 71 GWd/tM~~

~~200 400 600 800 1000 1200 Irradiation time (days)~~

~~Fig. V.2. Power burnup& ofGd&U rods irradiated3 BR n i~~

~~oo ABBREVIATIONS~~

~~. 1 Technical terms~~

~~a thermal expansion coefficient (unless otherwise mentioned) or annum, for year (used in units)~~

~~AO axial offset~~

~~ARO all control rods out~~

~~AZR axially zoned reactivity (fuel)~~

~~BA burnable absorber~~

~~BAF burnable absorber fuel~~

~~bcc body-centred cubic~~

~~BOC beginning of cycle~~

~~BPRA burnable poison rod assembly~~

~~C ban f controko l rods operating simultaneouslr yo calculated (hen used in conjunction with M)~~

~~Cp specific heat capacity~~

~~CPS control and protection system~~

~~D bank of control rods operating simultaneously~~

~~PWRfuea U measurn a r i l fo ) m f reactivitpc eo 0 50 $ x y(~~

~~DC direct current~~

~~DNB departure from nucleate boiling~~

~~DTC Doppler temperature coefficient~~

~~E Young's modulus~~

~~è creep rate~~

~~EDTA ethylene diamine tetraacetic acid~~

~~EFPD equivalent full power days~~

~~EDX energy dispersive X-ray~~

~~EFPM equivalent full power months~~

~~f cyclo d e en EOC~~

~~185 EOL end of life~~

~~FA fuel assembly~~

~~FGR fission gas release~~

~~FP fission product~~

~~FQ total peaking factor (sometimes calle t channedho l factor)~~

~~FR fuel rod~~

~~Fxv radial FR peaking factor at average elevation (mid-core usually)~~

~~FN x* nuclear contribution to FXY~~

~~G shear modulus~~

~~GAIN international data acquisition programme on the thermal-mechanical behaviour of Gd fuel to high burnups~~

~~GAP international data acquisition programme on the kinetics of Gd depletion in a Gd FR~~

~~R F d G gadolinia-urania fuel rods~~

~~GW(e) electrical power expressed in gigawatts~~

~~HBWR Halden boiling water reactor t fulho l poweHFr P~~

~~HLW high level waste~~

~~HZP hot zero power~~

~~IDR integrate routy ddr e (U02 conversion process)~~

~~k thermal conductivity~~

~~keff effective multiplication factor~~

~~IFBA integral fuel burnable absorber~~

~~KOFA Korean Fuel Assembly~~

~~LHGR linear heat generation rate, expresse kW/n di mr W/co m~~

~~LLLP low leakage loading pattern~~

~~LOCA los f coolanso t accident~~

~~LTA lead test fuel assembly (demonstration FA)~~

~~186 M measured~~

~~MOX mixed oxide U02-Pu02~~

~~MTC moderator temperature coefficient~~

~~MTR material testing reactor~~

~~MWd/tU initiallU burnu t r ype p containe d expresse x fuee W th lM dn i s da~~

~~v Poisson's ratio~~

~~NPP nuclear power plant~~

~~NSRR Nuclear Safety Research Reactor~~

~~OD outer diameter~~

~~o/m oxygen to heavy metal ratio~~

~~PC power coefficient~~

~~PCI pellet-cladding interaction~~

~~PIE post-irradiation examination~~

~~PCIOMR Pre-conditioning Interim Operating Management Recommendations~~

~~m pc reactivity expresse k( ,/k^o 10 n d)i f 6f 5 m pp part millior spe n~~

~~OC quality control~~

~~p density (g/cm3) clusted ro RCCr controA l assembly~~

~~R&D research and development~~

~~REE rare earth element~~

~~RIA reactivity initiated accident~~

~~a neutron cross-section~~

~~SEM scanning electron microscopy~~

~~SDM shut-down margin~~

~~UFR U02 fuel rod~~

~~UGF urania-gadolinia fuel (Russia)~~

~~TD theoretical density (g/cm3)~~

~~187 TGA thermogravimetric analysis~~

~~U fuel UO2 fuel~~

~~WOCA world outside centrally planned economy area~~

~~d G o weighw/ t percen Gd- t0 U0 Gd n i 0 2 2 3 2 3~~

~~WWER water water energy reactor (Russia typR en PW reactor )~~

~~y Zr Zircaloy~~

~~YGN Yonggwang NPP~~

~~2. Organizations~~

~~AtoB (SWEB mAB A ) B AB~~

~~ARSRIIM All-Russian Scientific Research Institute of Inorganic Materials (RUS)~~

~~N B BELGONUCLEAIRE (BEL)~~

~~BNFL British Nuclear Fuels, pic (UK)~~

~~BWFC Babcoc Wilcod kan x Fue (USAy C l )~~

~~CE Combustion Engineering ABB/Cw no , E (USA)~~

~~CEA Commissaria l'Energià t e Atomique (FRA)~~

~~EOF Electricité de France (FRA)~~

~~ENUSA Empresa Naciona l Uranide l (SPAA oS )~~

~~FBFC Franco-Belg Fabricatioe ed Combustiblu nd e (FR BELA& )~~

~~FRAGEMA Framatome-CogemE GI a (FRA)~~

~~GE General Electric (USA)~~

~~JAERI Japan Atomic Energy Research Institute (JPN)~~

~~JNF Japan Nuclear Fuel (JPN)~~

~~KAERI Korea Atomic Energy Research Institute (ROK)~~

~~KNFC Korean Nuclear Fuel Company (ROK)~~

~~KWU Kraftwerk Union, a group within SIEMENS AG (GFR)~~

~~MNF Mitsubishi Nuclear Fuel (JPN)~~

~~NE Nuclear Electric (UK)~~

~~188 NFI Nuclear Fuel Industries (JPN)~~

~~PNC Power Reacto Nuclea& r r Fuel Development Corporation (JPN)~~

~~RRC/KI Russian Research Center, Kurchatov Institute (RUS)~~

~~SCK/CEN Stuiencentrum voor Kernenergie/Centre d'Etude Nucléaire (BEL)~~

~~SPC Siemens Powe , previouslCy r (USAF yAN )~~

~~NucleaS USNRU r CRegulator y Commission (USA)~~

~~Wh Westinghouse (USA)~~

~~189 PARTICIPANT CO-ORDINATEE TH SN I D RESEARCH PROGRAMME~~

~~Bairiot. H , FEX, Belgium~~

~~Chantoin, P. International Atomic Energy Agency~~

~~Cho, N.Z. Korea Advanced Institut Scienceof Technologyeand , Republicof Korea~~

~~Delbrassine, A. Centre d'Etude l'Energie sd e Nucléaire, Belgium~~

~~Farrant, D. British Nuclear Fuels, United Kingdom~~

~~Gündüz, G. Middle East Technical University, Turkey~~

~~Hron, M. Nuclear Research Institute, Czech Republic~~

~~Onufriev, V.D. Scientific and Research Institute of Inorganic Materials, Russian Federation~~

~~Proselko. vV Kurchatov Institute of Atomic Energy, Russian Federation~~

~~Boulier. B , EDF/SEPTEN, France~~

~~Sukhanov, G. International Atomic Energy Agency~~

~~Toba, M. Nuclear Fuel Industries Ltd, Japan~~

~~191 QUESTIONNAIR IAEA-TECDOCN EO s~~

It would greatly assist the International Atomic Energy Agency in its analysis of the effectiveness Technicalofits Document programme could ifyou kindly answer followingthe questions and return the form to the address shown below. Your co-operation is greatly appreciated.

~~Title: Characteristics and use of urania-gadolinia fuels Number: IAEA-TECDOC-844 obtaiu yo d n di thi w s 1TECDOCHo . ?~~

~~[ ] From the IAEA: [ ] At own request ] Withou [ t request [ ] As participant at an IAEA meeting ] [ From professionaa l colleague [ ] From library rat u contene eyo TECDOCe th o th d f to w 2Ho . ?~~

] Useful[ , includes informatio t founnno d elsewhere [ ] Useful as a survey of the subject area [ ] Useful for reference [ ] Useful because of its international character ] Usefu[ traininr fo l studr go y purposes [ ] Not very useful. If not, why not? becomu yo o d e w awarTECDOCe 3Ho .th f eo s available fro IAEAe mth ?

] Fro[ m reference: sin [ ] IAEA publications ] [Othe r publications [ ] From IAEA meetings ] [ From IAEA newsletters othey B r mean] [ s (please specify) finu t difficuldi yo f I obtai o t ] t [ n informatio TECDOCn no s please tick thix sbo

~~4. Do you make use of IAEA-TECDOCs?~~

~~[ ] Frequently [ ] Occasionally [ ] Rarely~~

~~5. Please state the institute (or country) in which you are working:~~

~~Please return to: R.F. Kelleher Head, Publishing Section International Atomic Energy Agency 0 10 x P.OBo . Wagramerstrasse 5 A-1400 Vienna, Austria~~