oxidation

207

APPENDIX2

THETHERMALOXIDATIONOFGRAPHITE

ard

l

Graphiteisudinavarietyofnuclearreactortypes;principallyformoderator,reflector,fuel

tisaformofcarbon,likecoalandcharcoal,itsoxidation

dising

environmentsofparticularinterestareair(oxygen),carbondioxideandsteam(water).

onsandThermodynamics

onwithOxygen

½O

2

+C=CO∆H=--1(A2.1)

O

2

+C=CO

2

∆H=--1

(A2.2)

where∆Histhestandardenthalpyofformationat298°C.

Reaction(A2.1)maximistheamountofcarbonwhichmayberemovedbyagiven

massofoxygen(ascarbonmonoxide).Reaction(A2.2)maximistheamountofheat

producedbyoxidisingagivenmassofcarbon(tocarbondioxide).Reaction(A2.2)mayalso

beregardedasproceedinginstages,withreaction(A2.1)followedby

½O

2

+CO=CO

2

∆H=--1

(A2.3)

(obtainedbydifference-Hess’slaw)

Reaction(A2.3)cantakeplacewhollyinthegaspha.

ethisfact,

puredennucleargraphitesdonotreadilyreactwithair,sokineticfactorsareobviouslyof

importance.

onwithCarbonDioxide

Thiscanbereadilyinferredbymultiplyingreaction(A2.1)by2,includingthe∆Hterm.

Theresultisthenaddedtoreaction(A2.2)l,the(Boudouard)reactionis

C+CO

2

=2CO∆H=+-1(A2.4)

onwithWater(watergasreaction)

C+H

2

O=CO+H

2∆H=+-1(A2.5)

also

208

C+2H

2

O=CO

2

+2H

2∆H=--1(A2.6)

Thehydrogenproducedcanthenreactwithcarbon:

C+2H

2

=CH

4∆H=--1(A2.7)

Subtracting(A2.5)from(A2.6)gives

CO+H

2

O=CO

2

+H

2∆H=--1

(A2.8)

Thisisthewatergasshiftreactionwhichtakesplaceinthegaspha.

Inmanypracticalcas,productsfromtheabovereactionsarefreetoescape,suchthat

halpychangesareofimportance,however,

sincetheygiveameasureoftheheatproducedinexothermicreactions.

isms,RegimesandKinetics

isms

T-catalyd

oxidationtypicallyfollowstheroute:

(i)Transportofoxidanttothegraphitesurface.

(ii)Adsorptionofoxidantontothegraphitesurface(physisorption).

(iii)Formationofcarbon-oxygenbonds(chemisorption).

(iv)Formationofcarbon-hydrogenbondsinreaction(A2.7)(reduction).

(v)Breakingofthecarbon-carbonbonds.

(vi)Desorptionofcarbonmonoxide,orotherproduct.

(vii)Transportofreactionproductfromthegraphitesurface.

Anyoftheabovestepsmayberatecontrolling,pthemajorreactant

concentrationgradient.

llingFactors

Factorscontrollingtherateofoxidationmayincludethefollowing:

(viii)Therateatwhichtheoxidantissuppliedtothesurface.

(ix)Thepartialpressureoftheoxidant.

(x)Thereactivesurfaceareaavailabletotheoxidantatthesurface.

209

(xi)Theamountanddistributionofcatalyticimpuritiesinthegraphite.

(xii)Thetemperature.

(xiii)Therateatwhichreactionproductsareremoved.

(xiv)Thefastneutrondamagetothegraphite.

(xv)Theamountofpre-oxidation(radiolyticorthermalburn-off).

(xvi)Thequantityofin-poredeposits.

(xvii)Theeffectivediffusioncoefficient.

ionRegimesandKinetics

Regime1

Atlowoxidationrates(generallyatrelativelylowtemperaturesfortheparticularoxidation

reaction)theoxidantmaybeatesntiallythesameconcentrationthroughoutthetransport

‘chemical’regimeischaracteridbythefactthatthereactionrate

islargelydeterminedbytheintrinsicreactivityofthegraphite(steps(ii)to(vi),above).

Differentpartsofthestructuremayreactatdifferentrates;thebinderbeingmorereactive

thanthegristparticlesandedgeatomsbeingmorereactivethanbasalplaneatoms,for

lsobethecathatthegascompositionvariesinnon-transportporesand

thattheeffectofagivengascompositionvarieswithporeshape(becauofgaspha

reactions).

Reactionbetweenairandpurenucleargraphitesisgenerallynotmeasurablebelow

about350°Candonlybecomessignificantintheregionof400°eofreactionis

-1.s-1atthistemperature(historicallyexpresdin

µg/gh;1µg/-1.s-1).Thechemicalregimethenextendstypicallyup

to550-600°C.

Theunitsforoxidationrateimplyaratelawoftheform

dm

dt

km=(A2.9)

where

m=graphitemass(kg)

t=time(s)

k=a(rate)constant(s-1)

However,forasolidreactingbody,ratelawsoftheformshownbelowwouldbe

expectedforreactionatthesuperficialsurface(withsomesimpleassumptions):

Slabs

dm

dt

k=(A2.10)

210

Cylinders

dm

dt

km=1

2(A2.11)

Spheres

dm

dt

km=2

3(A2.12)

Thereactionis,ofcour,ot,

however,berelatedtotheinitialopenporevolume(whichmightbeexpectedtobe

proportionaltographitemass)sincethereactionmusttakeplaceatthesurfaceofthepores.

Theratelawisthusonlysuperficiallysimilartoahomogeneousfirstordergaspha

expressionandcaremustbetakeninitsu,particularlyindescribingthetimedependenceof

graphiteburnoff.

Thevariationofreactionratewithtemperatureisofimportance:

kAe

E

RT=−

(A2.13)

where

Aisapre-exponentialfactor(s-1)

Eistheapparenttemperaturecoefficientofreaction(‘activationenergy’)(-1)

Risthegasconstant(-1.K-1)

-1.

Similarconsiderationsapplytoreactionwithcarbondioxideandwatervapour(H

2

O).

ThereactionwithCO

2

isoflessimportance,however,sinceitisnegligibleat625°C

(Thurlbeck,1962)anddoesnotpoaproblemevenatthehighestAGRinnersleeve

temperaturesof675°C(Prince,1976).Thereactionisalsoendothermic(equation(A2.4))

andsodoesnothavethesamesafetyimplicationsasthereactionwithair.

ThereactionwithH

2

OisofparticularimportanceinHTRreactors,becauofthe

(generallysmall)inleakagefromthesteamsideintothegascircuit,whereitcanreactwith

hepartialpressureofthewatervapourisavariableinthis

system,rateequationsoftheform

rAPn=(A2.14)

areapplied,where

r=thespecificreactionrate(-1.s-1)

A=a(rate)constant(s-1.(N.m-2)-1)

P=partialpressureofwatervapour(N.m-2)

Thereactionwithwatervapourisgenerallyinsignificantbelow800°Cand

approximatelyobeyquation(A2.14)withn=0.5overthetemperaturerange1000-1200°C.

ThekineticscanalsobedescribedbyaLangmuir-Hinshelwoodscheme(Walkeretal,1959;

Atkins,1987;Stairmand,1990).

Regime2

Inthisregime,thereactionratebecomeshighenoughforaccessofthegastothein-pore

211

structuretobesignificantlylimitedbydiffusioncontrol(steps(i)and(vii),above).Thiscan

air-graphitesystem,thisoccurs

approximatelyintherange600-900°‘activationenergy’ishalvedinthisregimeand

thekineticexpressionsinvolvetheeffectivediffusioncoefficientforthegraphite(Walkeret

al,1959;GibersonandWalker,1965).

Regime3

Thisisthemasstransferregime(Burnetteetal,1979;RaederandGulden,1989),where

reactionatthesuperficialsurfaceofthegraphiteissohighthatmostoftheoxidantis

consumedthere,theoxidantconcentrationgradientgenerallydevelopingacrossthelaminar

ctionrateisnowexpresdintermsofthesuperficialsurfaceareaofthe

graphite(kg.m-2.s-1)ngefrom

oneregimetoanothermaybeprogressiveandmode2mayappeartobemissinginsome

cas.

Twoother‘regimes’appeartrivial,butcanufullybedistinguished:

Regime4

Ifthereisafixedrateofingressofoxidanttothesystem,forexampleasaknownquantityof

impurityinthemake-upgas,therateofoxidationcannotexceedtherateofsupplyofoxidant

(ratebalance).Thepreferredsiteofanyresultingoxidationmaynotbeknown,however.

Regime5

Ifthesystemcontainsafixedamountofoxidant,forexamplethatremainingafterblowing

downandrechargingthecoolantgas,theextentofoxidationislimitedbytheamountof

oxidantavailable(massbalance).Neitherthelocation,northerateofreaction,maybe

knowninthisca.

sis

Thekineticsinthechemicalregimemaybefurthercomplicatedbycatalysis(McKee,1981).

Thecatalyst(impurity)particlesacttoincreathereactionratebyofferinganalternative

onratesareparticularlyincread

atlowertemperatures(thisleadstoa‘compensation’effect).

Asimplecatalyticmodelinvolvestheoxidationofmetalatoms(M)byoxygen,

followedbyreductionoftheoxidebycarbon:

½O

2

+M=MO(A2.15)

MO+C=CO+M(A2.16)

Thereactionmayproceedbya‘tunnelling’mechanism.

Reactioninhibitorsarealsoknown(McKee,1991)andsomesubstancesareabletoact

toeitherpromoteorinhibitreaction,perhapsbycompetingforactivesites(e.g.

water/oxygen).Boron(withphosphorus)isknownasaninhibitorforthermaloxidation,but

whenintercalatedintothegraphitestructurewillpromoteoxidation(Karraetal,1995)

212

(perhapsbymimickingfastneutrondamage).

ementsfortheInformation

Earlydesignsofgraphitemoderatedreactoroperatedwithairasthecoolantandtherewasa

requirementtounderstandboththelikelyongoingoxidationbehaviourofthegraphiteandits

ereactorsweresuperdedbycarbon

dioxidecooleddesigns,anunderstandingofthereactionofgraphitewithairremained

importantforthefollowingreasons:

(i)Safetycainformationrelevanttobothmajorandminoringressofairtothesystem

underfaultorotherconditions(Dodson,1960;NairnandWilkinson,1960;Blanchardand

Fitzgerald,1978).

(ii)Thepossiblerequirementstocarryoutdeliberateoxidationsto

(a)Removedepositsfromfuel.

(b)Openupthestructureoflowdiffusivitygraphitetoimproveinhibitor(e.g.

methane)access.

(c)Removedepositsfromfuelpinandheatexchangersurfacestoimproveheat

transfer.

(d)Estimatetheamountofdepositingraphitemoderatorandfuelsleevesby

differentialthermaloxidation(Welch,1972;OxleyandDymond,1972;BaguleyandLivey,

1972)(soastobeabletocorrecttheweightloss).

(e)Alterthestructureofexperimentalgraphitesincontrolledwaystoimprove

theoreticalknowledgeoftheinteractionbetweenreactivityandstructure.

ialMeasurementsandKnowledge

Thecomplexityofgraphiteoxidationbehaviourissuchthatthefollowingaregenerally

required:

(i)Goodstatisticaldataontherelevantoxidationratesforarangeofblocksandheats.

(ii)Anawarenessoftherelevantoxidationregimeandtheratelawswhicharelikelyto

apply.

(iii)Informationonthetemperaturecoefficientofreaction.

(iv)Informationormeasurementontheheatchangeonreaction.

(v)Theoreticalorexperimentalinformationonfactorsaffectingtheoxidationrate,suchas

fastneutrondamage,burn-up,depositionofpotentialcatalysts,etc.

(vi)Modellingknowledgetoextrapolatefromsmallscalesamples,orfullscaletests.

Assomegraphitereactorsarecomingtotheendoftheirlives,thereisanincreasing

requirementtocarryoutasssmentsforlongtermstorageordisposal(Wickhametal,1996).

Thepotentialoxidationbehaviourisalsoofconcerninthiscontext.

ledgements

Incompilingthisaccount,theauthorhasbeenparticularlygratefulforreviewsbyJohn

Stairmand(Stairmand,1990)andTonyWickhametal(Wickhametal,1996)whointurn

(Walkeretal,1959)andDubinin(Dubinin,1966).

Anyimportantomissionsaretheauthor’sown,however.

213

REFERENCES

[1]alChemistry,3rdedition,OxfordUniversityPress,783(1987).

[2]liedtoGraphite/CarbonDepositSystems-Paper

11,TheAttackRateoftheunderlyingGraphiteduringOxidationofIn-poreDeposits,

SymposiumonDifferentialThermalOxidationheldatBerkeleyNuclearLaboratories

(1972).

[3]ReactivityofWAGRModeratorandSleeve

Graphites,UKAEAReportWAGR/TC/P(78)21(1978).

[4]BurnetteR.D.,softheRateofSteamOxidation

ofGraphiteatHighGasVelocity,14thBiennialConferenceonCarbon,Pennsylvania

StateUniversity(1979).

[5]thofOxidationofGraphite:ATheoreticalApproach,UKAEA

ReportDEG148(CA)(1960).

[6]tryandPhysicsofCarbon,2(,Jr),MarcelDekker,

NewYork,51(1966).

[7]onofNuclearGraphitewithWaterVapour-Part

1:EffectofHydrogenandWaterVapourPartialPressures,BattelleNorthwestReport,

BNSA-181(1965).

[8]KarraM.,ZaldivarR.J,RellickG.S.,tutional

BoroninCarbonOxidation-InhibitororCatalyst?,Proc.22ndBiennialConferenceon

Carbon,SanDiego(1995).

[9]tryandPhysicsofCarbon,16(r,Jr),MarcelDekker,

NewYork,1(1981).

[10]tryandPhysicsofCarbon,23(r,Jr),MarcelDekker,

NewYork,173(1991).

[11]dictionofConditionsforSelf-sustainingGraphite

CombustioninAir,PaperGCM/UK/11,US/UKCompatibilityConference(1960).

[12]liedtoGraphite/CarbonDepositSystems-Paper8,

DTOStudiesofIrradiatedGraphiteusingGasChromatography,Symposiumon

DifferentialThermalOxidationheldatBerkeleyNuclearLaboratories(1972).

[13]enceintheDesignofGraphiteModeratorStructures,Paperprentedto

theNuclearEngineeringSociety(1976).

[14]AnalysperformedbyNET,IAEATechnical

CommitteeMeetingonFusionReactorSafety,Jackson,Wyoming(1989).

[15]teOxidation-ALiteratureSurvey,AEATechnologyReport

AEA-FUS-83(1990).

[16]idCompatibilityReviewofReactorMaterialsforCO

2

-cooled

GraphiteModeratedReactors,TRGReport267(W)(1962).

[17]WalkerP.L.,esinCatalysis,11,123(1959).

[18]liedtoGraphite/CarbonDepositSystems-Paper1,Principlesand

Problems,SymposiumonDifferentialThermalOxidationheldatBerkeleyNuclear

Laboratories(1972).

[19]WickhamA.J.,MarsdenB.J.,sibilityand

ConquencesofGraphiteCoreDegradationduring‘CareandMaintenance’and

‘Safestore’,AEATechnologyReportAEA/16423595/R/001(1996).

214