ML101530556

Mps GL-04-002 Draft RAI Responses/Public Telecon Handouts

ML101530556
Person / Time
Site:	Millstone
Issue date:	06/02/2010
From:	Office of Nuclear Reactor Regulation
To:
Sandeers, Carleen, NRR/DORL, 415-1603
References
GL-04-002, FOIA/PA-2011-0115
Download: ML101530556 (54)

Text

Page1of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

0BMillstonePowerStationUnit2(MPS2),RAI3 DNC'sresponseisunclearastohowlikelyitisthatthestreamofbreakflowwillbebrokenup.Basedon theMPS2audit,theNRCstaffbelievesthatasignificantportionofoneofthestrainerarraysislocatedin aloopcompartmentbeneathpipingsubjecttobreaking.Withoutacoverplate,itisdifficulttoconclude thatliquidfallingfromthebreakwouldnotfallintothecontainmentpoolabovethestrainerarraywith sufficientkineticenergytoresultinairentrainment.Also,theNRCstaffnotesthatpage7ofAttachment 1,totheDecember18,2008,DNCletterstatesthat"...manyofthepossiblebreaklocationsareabovea portionofthestrainerandbreakflowintheseareaswouldkeeptheportionofthestrainerbelowthe breakclearofdebris."ItisnotcleartotheNRCstaffwhyairentrainmentwouldnotoccurifmanyofthe breaksresultinwaterfallingfromthebreakontothestrainersuchthattheaffectedportionofthe straineriscontinuallyclearedoff.Also,theflowcontrollingbafflesinsidethestrainermayencourage uniformflow,butwhenenergeticwaterissplashingdownontoastrainerarrayfromabove,itisnot clearhowthebafflecanlimittheairentrainmenttoanegligiblequantity.Itisnotclearthatthestrainer bafflesweredesignedtocompensateforsuchanonuniformexternalflow.Pleaseclarifythesepoints.

Thebasisfortheclaimthatairwillescapefromthestrainerfinsisnotclear.Basedonthedescriptionin theresponsesforMPS2,itisnotclearwhythe11/2inchopeninginthetopofthestrainerwould performdifferentlythantherestofthestrainer,orwouldnotbecoveredwithdebris,justlikeanyother strainersurface.Pleaseclarifythesepoints.

RegardingtheFroudenumberdiscussion,thebasisforthedeterminationthataircouldnotreachthe suctionpipesbasedontheFroudevaluewasnotclear.Oneparticularpointthatwasunclearconcerned theassumedsizeoftheairbubblesandwhethertheFroudenumberlimitreferredtowasassociatedwith vortexingorbubbleingestion.Pleaseprovidethebasisorreferenceusedforthisassumption.MPS2cites theheadlosstestsperformedbyAtomicEnergyofCanada,Ltd(AECL).SomeoftheAECLheadlosstests experiencedairinthepumpsuctionline.PleaseaddresshowthisimpactsMPS2'sevaluationofsump performance.PleaseaddresswhethertheFroudenumberwasexcessiveforthesetests(e.g.,greater than0.31).Ifthereisdirecttestingevidencethatcouldhelpresolvethequestion,pleaseprovidesuch documentation.

BasedontheNRCstaff'sunderstanding,anyairingestedbythestrainerwouldseeminglyremain trappedinside,accumulatinguntilitwasabletoexitthroughtheperforatedplateorintothesuction lines.AiringestioniscomplexanditisuncleartotheNRCstaffwhichwaytheairwouldeventuallygo andhowmuchwouldaccumulateinthestrainerbeforesteadystateconditionsarereached.The installationofacoverplatewouldpreventwaterfromsplashingdownontoandentrainingairintothe strainer,removingsomeofcomplexissuesassociatedwithairingestion.

Pleaseaddresstheaboveairingestionconcernsbecauseexcessiveairingestioncandegradeoperationof thepumps,whichtakessuctionfromthesump.

Page2of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

7BResponsetoMPS2,RAI3 TominimizethelikelihoodforairingestionDNCwillinstallasafetyrelatedcoverplateduringthe2R21 refuelingoutage.Theproposedcoverplatewillcovertheentirestrainer.Thiswillpreventwaterfrom splashingdownontoandentrainingairintothestrainer.Thecoverplatedesignuses3/16inchthick stainlesssteeldiamondplate.Belowthediamondplateis13 4 inchstainlesssteelgrating.Thebottom surfaceofthecoverplateisatleast9inchesabovethestrainerfinstoavoidencapsulation.Testingwith thefulldebrisloadandthecoverplatewasdonewiththebottomofthecoverplate9inchesabovethe topsofthefinstoavoidstrainerencapsulationbydebris.Thecoverplatedoesnotoverhangthe strainer.Theinitialdesignestablishedtheloadbearingcapacityofthecoverplateas60lb/ft2.

Evaluationsofthefinaldesignloadingforthecoverplatewilltakeintoaccountthewaterfallfromthe postulatedbreak.

1BMPS2,HeadLossandVortexing,RAI6 ThisRAIidentifiedsomedifferencesinnonchemicalheadlossesbetweenthetwotestfacilities.The December18,2008,DNCletterprovidedthefirstdocketedinformationprovidingsignificantinformation ontheMPS2Rig89testing.TheNRCstaffhasreviewedthisanddeterminedthatthefollowing informationisneededtocompletethereview:

a. PleaseprovideinformationthatjustifiesthattheRig89headlosstestwasconductedwitha fibrousdebrisloadthatmaximizednonchemicaldebrisheadloss.

b. Pleaseprovideinformationregardingwhetherthedebrisbedcontainedadequatefibertoensure thatamaximumheadlosswasattainedwithoutbeddisturbanceslimitingtheheadloss.

8BResponsetoMPS2,HeadLossandVortexing,RAI6,Issuea FortheMPS2Rig89chemicaleffectstest,thethinbeddebrisadditionmethodologywasusedto maximizethenonchemicaldebrisbedheadloss.Thefullparticulatedebrisloadwasaddedatthestart ofthetest,andthenadditionsoffibrousdebrisweremadein1/16inches(1.6mm)theoreticalbed thicknessincrements.Notethatthetheoreticalbedthicknessisdefinedastheuncompressedfiber volumedividedbythetestmodulesurfacearea.Thefirstfiberaddition(1/16inches(1.6mm))was made30minutes(enoughtimefordebrispreparation)aftertheadditionoftheparticulatedebris.The secondfiberaddition(anadditional1/16inches(1.6mm))wasmade30minutesafterthefirstaddition.

PreviousthinbedtestsconductedinRig33(reducedscale)[D1D]andRig42(largescale)[D2D]had determinedathinbedthicknessof1/8inches.Rig33testresultswereusedtoestablishathinbed.In bothofthereportedrig33thinbedtests,threefiberadditionsweremadetoverifythatthethirdfiber additiondidnotresultinanysignificantheadlosschange.Sincethethirdfiberadditioninbothofthese testsresultedinnosignificantadditionalheadloss,thethinbedwasdeterminedtobe1/8inch.The debristostrainersurfacearearatiosinTable1belowreflectthreefiberadditionsintherig33testsand twofiberadditionsintherig89test.Athirdfiberadditionintherig89testwasunnecessarysincethe thinbedthicknessof1/8wasestablishedinrig33.Thefiberdensitywas2.4lb/ft3.

Page3of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Theparticulatedebrisloadinthereducedscalethinbedtests(TestsM222andM227)wasgreater thanintheRig89test.IntheRig89test,anupdateddebrisloadwasusedbasedonDominion EngineeringTransmittal25203ER070029Rev0[D3D].XTable1Xliststhedifferenttestdebrisquantitiesper unitteststrainersurfaceareaforRigs89and33.Theparticulatedebrisquantitywasalmosttwotimes greaterinRig33teststhanintheRig89test.

AtthetimeofMPS2Rig89testing,itwasbelievedthattheteststrainerareawas5.74ft².Aftertesting, however,itwasfoundthattheRig89finareahadbeenmiscalculated[D4D].ThetrueRig89finareawas recalculatedtobe5.08ft².Therefore,thetestedfibrousdebrisamountwasalmost13%morethan intended.Theextrafibrousdebrismightsettleonthetesttankfloororattachtothedebrisbedloosely.

Table1:MPS2TestDebrisLoadComparisonsbetweenRigs89and33 Rig WalnutShell[lbm/ft2]

Nukon[lbm/ft2]

Knauf[lbm/ft2]

MineralWool[lbm/ft2]

89 0.38 0.018 0.028 0.016 33 0.69 0.024 0.037 0.020

AttheendoftheRig89test,itwasfoundthattheamountofdebrisattachedtothefinsurfacewas

~56%,whichisquitecomparabletothedebrisonthefinforRig33tests(~58%and66%).InRig33tests itwasdemonstratedthatthechangeinheadlossafter1/8inchesfiberadditionwasinsignificant.Since Rig89testhadaboutthesameamountoffiberonthefinswithnearlyhalfoftheparticulatequantity perunitareaofthefins,theamountoffiberinthedebrisbedshouldbeequalormorethanthe requiredquantityoffibertoformathinbed.

Figure1Xshowsthedebrisbedafterthetest.Itcanbeseenthatauniformdebrisbedwasformedacross thewholesurfaceoftheteststrainer.XFigure2Xshowsaportionofthedebrisbedremovedfromthe strainersurfaceafterthetest.

Page4of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure1:DebrisBedaftertheTestM2C1

Page5of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure2:CloseupofDebrisBedafterRemovingfromFin

9BResponsetoMPS2,HeadLossandVortexing,RAI6,Issueb Asexplainedabove,thefibrousdebrisadditionsintheRig89MPS2testweresufficienttoformathin bedandtohavemaximizedthenonchemicalheadloss.Ifmorefiberwereadded,theextrafiberadded tothetestwouldeitherlooselyattachtothethinbedsurface,formingaporouslayer,orsettlebetween andinfrontofthefinsonthetankfloor.

Figure1XandXFigure2Xalsoshowthedebrisbedwasfirmanduniformacrosstheentiresurfaceofthefin withnocrackorholeorotherdegradations;alsothefinalthicknessofthebedwasmeasuredtobe nearly1/4inches,whichwasmorethanthetheoreticalthinbedthickness.Thefullydevelopedthinbed thicknessthusensureddebrisbedstructuralintegrityforthesubsequentchemicaladditions.

Flowsweepswereperformedattheendofthetest.Thechangesinheadlossduringtheflowsweeps showednosignsofhysteresis,andheadlosschangeswerereversible.Aftertestdebrisbedexamination didnotfindanysignofdegradations,suchaslargeholes,dislocationsorfractures.Thus,thedebrisbed wasnotdegradedduringthetestandheadlosswasnotlimitedbyholesin,ordislocationof,portionsof thedebrisbed.

Page6of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

2BMPS2ChemicalEffectsQuestions FollowingreviewofthechemicaleffectsevaluationdetailsintheDNCDecember18,2008,letter,the NRCstaffidentifiedthatthefollowingadditionalinformationwasneededinordertodetermineifthe testingwasperformedinanacceptablemanner:

12.TheMPS2calciumdissolutiontestatapHof7.0resultedina30daycalciumconcentrationof126 mg/L.DNC'sDecember18,2008,letterstatesthatthepH7.0case(withouttrisodiumphosphate present)wasusedtodeterminetheconcentrationofcalciumintheRig89test.However,thecalcium concentrationusedforRig89testingwas40.4mg/L.Pleasejustifywhy40.4mg/Lisarepresentative valueintheRig89testingwhenthedissolutiontestingconductedwithscaledquantitiesofconcrete resultedinacalciumconcentrationof126mg/L.

13.InAttachment1,Table02,ofDNC'sDecember18,2008,letter,thecalciumconcentrationfortime infinityisshownas117mg/LforpH7.0.Pleaseexplainwhythisconcentrationfortimeinfinityis appropriate,giventhe30daybenchtestcalciumconcentrationatpH7.0was126mg/L.

14.DNC'stestingwasperformedat104F,whichiswellbelowearlypostIossofcoolantaccidentpool temperatures.Thesolubilityofcalciumphosphate(hydroxyapatite)decreasesasthetemperature increases.PleasediscusswhethermorecalciumphosphateprecipitatewouldhaveformedintheRig89 testsifthistestwouldhavebeenperformedathighertemperature.Ifmorecalciumphosphate precipitatewouldbeexpectedatahighertemperature,whentheshorttermNPSHmarginisapplicable, pleasejustifywhytheoverallRig89testresultsprovideforanadequateevaluationofchemicaleffects.

15.PleasecomparethetotalamountofaluminumthatispredictedtobereleasedbytheAECLmodel withthatpredictedbytheWCAP16530basemodel(i.e.,norefinementsforsilicateorphosphate inhibition).Discussanysignificantdifferencesbetweentheplantspecificpredictionsforthetwo methods,includingtheacceptabilityofthesedifferences.

10BResponsetoMPS2,ChemicalEffectsQuestion12 Thevalueof40.4mg/LusedintheRig89testingwascalculatedbyappropriatelyscalingtheresultsof thedissolutionteststomatchupdatedestimatesoftheMPS2concretesurfacearea.Thisresponsewill show:

1. Theconcretesurfaceareatovolumeratiousedinthebenchtopdissolutiontestswasbasedon estimatesoftheconcretesurfaceareathatwerelaterupdated;

2. Theresultsofthedissolutiontestsmaybenormalizedtounitsofcalciumreleaseperunitarea, whichmaythenbeusedtocalculatetheexpectedcalciumreleaseandcalciumconcentrationin MPS2basedontheupdatedconcretesurfacearea;

3. Itisappropriatetousethefittotheentiredatasettodeterminethescaledcalcium concentrationratherthantoscaletheanalysisresultobtainedonday30(126mg/L),whichis moresubjecttosamplingandstatisticalerrors.

Page7of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Theconcretesurfaceareatovolume(SA/V)ratiousedinthebenchtopdissolutiontestswasroughly3 timesgreaterthanthecurrentcalculatedSA/VratiousingdatafromERC25203ER060007Rev.3[D5D]

andleadstotheapparentdiscrepancy.ThedissolutiontestsconductedfromFebruarytoMarch,2008, usedcouponssizedtomeettheSA/VratiocalculatedfromRev.1ofthatdocument[D6D]andincluded scaledquantitiesoffibrousdebris.XTable2XcomparestheSA/Vratiousedinthedissolutionteststo thosecalculatedfromthesourcereferences.Itisimportanttonotethat,bydesign,thereisno uncoatedconcretewithintheMPS2containmentandthatallvaluesquotedareconservativeestimates ofbareareasexposedeitherbychippingandwearorbyimpactofthebreakjet[X5X].

Table2:ComparisonofDissolutionTestConcreteSA/VRatiotoMPS2Values Source ERC25203ER060007, Rev.1[X6X]

DissolutionTest ERC25203ER060007, Rev.3[X5X]

Date 2007/08 2008/02-2008/03 2008/04 SubmergedConcrete 3700ft2

[3.44x106cm2]

1.1x3.4x0.5cmcoupons

[11.98cm2]

400ft2

[3.7x105cm2]

ExposedConcrete 2600ft2

[2.42x106cm2]

925ft2

[8.6x105cm2]

Volume 41,000ft3

[1.16x106L]

4L 41,800ft3

[1.18x106L]

SA/VRatio(Submerged) 2.97cm2/L 3.0cm2/L 0.31cm2/L SA/VRatio(Total) 5.05cm2/L

1.04cm2/L

BecausetheconcreteSA/Vratioforcontainmentdiffersfromthattested,theresultsobtainedarenon representativebutmaybeappropriatelyscaled.Normalizationofthedissolutiontestdatamaybe performedbydividingtheresults(inmg/L)bytheSA/Vratio(3.0cm2/L),asindicatedbytherighthand verticalaxisinXFigure3X.Similarly,thefittothecalciumconcentrationdata,describedbelow,mayalsobe normalizedtoproduceacalciumreleaseequation.Thus,the30daycalciumreleaseperunitareaof concretecanbereadfromthefigureorcalculatedfromthefitandusedtocalculatethecalciumrelease fromaknownsurfaceareaofconcrete.

XFigure3XalsoshowsfirstordercurvefitstothedatarepresentedbyEquations1and2.Thesewere determinedusingrobustfittingprocedureswithinTableCurve2DF Fthatreducethefittingerrorscaused bydataoutliers.TheconstantsfoundwithinEquation1werereportedinTable25ofthebenchtop TestReport[D7D]andTableO2ofDNCsDecember18,2008letter.Equation2maybecalculatedfrom Equation1bydividingtheinitialconstantbythetestedsurfaceareatovolumeratio,3.0cm2/L.

Equation1

Equation2

TableCurve2DisproducedanddistributedbySystatSoftwareInc.

Page8of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure3:CalciumReleaseDatafromMPS2pH7andpH8DissolutionTestswithoutTri SodiumPhosphateat90C

Notethelinesarefitsofthedatasetstoafirstorderreleaseequation.

Itisappropriatetousethefitratherthantherawdatatodeterminethe30daycalciumconcentration, asdriftsinpH,samplingerrors,andstatisticalerrorassociatedwiththeanalysistechnique,ICPOES (InductivelyCoupledPlasmaOpticalEmissionSpectroscopy),mayalterthemeasuredconcentration.

After30days,theexpectedcalciumreleaseatpH7and90Cis:

UsingthetotalSA/Vratiofromthe4thcolumnofXTable2X,1.04cm2/L,theexpectedcalcium concentrationis

Forcomparison,theexpectedcalciumconcentrationatpH8is23.7mg/Lbysimilaranalysis.

ThisresultmaybecomparedtotheWCAP16530methodofcalculatingcalciumrelease,asdescribedby Laneetal[D8D].Inutilizingthismethod,thecalculatedpHhasbeenused;inordertomaximizethe releaserate,themaximumpHwasusedtocalculatethereleasefromNuKonandMineralWool(around Time (h)

Time (d)

[Ca] (mg/L)

Ca Release (mg/cm2) 0 96 192 288 384 480 576 672 0

2 4

6 8

10 12 14 16 18 20 22 24 26 28 30 0

20 40 60 80 100 120 140 160 180 200 0

8 16 24 32 40 48 56 64 pH 7 pH 8

Page9of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

pH8.3)andtheminimumpHwasusedtocalculatethereleasefromconcrete(aroundpH8.0).Bythis method,thecalculatedcalciumreleasefromconcreteisminisculeF1 F;mostofthecalciumreleasedcomes fromfibrousdebris.Thecalciumconcentrationispredictedtoplateauat12.3mg/L(XFigure4X),the saturationlimitofcalciumreleasedfromNuKonatpH8.3and189F(87.2C).Therefore,thecalcium concentrationobtainedbyscalingtheAECLpH7dissolutiontestresultsisconservativewithrespectto theWCAPresult.

Figure4:CalciumreleasefromMPS2fibrousdebrisandconcreteascalculatedbytheWCAP method[X8X]

1BResponsetoMPS2,ChemicalEffectsQuestion13 Aswithanydatum,themeasuredcalciumconcentrationfortheday30samplewassubjecttosampling andstatisticalerrors,withthemagnitudeofthestatisticalerroralonebeingabout+/-8mg/L.Whenthe dataareevaluatedasawhole,thedatafitproducedbyTableCurve2Dcalculatedacalcium concentrationof117mg/Lfortimeinfinity;itshouldbenotedthatthisvalueisarguablyidenticalto 126mg/Lwithintheexperimentalerror.Sourcesoferrorarefurtherdiscussedbelow.

ThedissolutiontestdataandthefirstorderfittothemarerepresentedinXFigure5X.

1WhenNukonandMineralWoolcontributionstocalciumreleaseareneglected,thecalculatedcalciumrelease fromconcreteusingtheWCAPmethodislessthan7g.Bycontrast,whenNuKonandMineralWool contributionsareincluded,thecalculatedcalciumreleaseisnearly15kg.

Time (h)

Time (d)

[Ca] (mg/L)

Ca Release (kg) 0 96 192 288 384 480 576 672 0

2 4

6 8

10 12 14 16 18 20 22 24 26 28 0

2 4

6 8

10 12 14 16 18 0

3 6

9 12 15 18 21

Page10of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure5:CalciumReleaseDatafromMPS2pH7andpH8DissolutionTests(withoutTSP)

Note:thelinesarefitsofthedatasetstoafirstorderreleaseequation.

Itisimmediatelyapparentthatthedatacontainafewoutliers.Itisimportanttoconsiderthateach datumsufferstosomedegreefromexperimentalerrors(positioninthesolutionatwhichthetubewas placed,contaminationofsamplevials,insufficientfilteringofsamplesbeforeanalysis,pHdrift,etc.).

Forexample,ifthesamplehadbeentakenfromalocationintheflasknearasourceofcalcium (concreteorfibrousdebris),theresultcouldhavebeenbiasedhigh.Thereisalsouncertainty(statistical error)associatedwiththeanalysistechnique,ICPOES;thisuncertaintywasapproximately+/-8mg/Lfor the117mg/Lmeasurement.Therefore,consideringallsourcesoferroranduncertainty,thereisno statisticallysignificantdifferencebetween117mg/Land126mg/L.

12BResponsetoMPS2,ChemicalEffectsQuestion14

Potentiallyincreasedcalciumphosphatesolubilityathighertemperaturesdoesnotsignificantlyimpact theMPS2testresultsduetosignificantconservatismsbuiltintothetestingprogram.

1. ThereisnosignificantsourceofcalciumintheMPS2containment.Theonlypotentialcalcium sourcesfortheMPS2containmentareuncoatedconcreteanddislodgedfibrousinsulation.By design,thereisnouncoatedconcreteintheMPS2containment.FortheRig89testing,atotal of1325ft2ofconcreteisassumedtobeuncoatedincontainment.Ofthattotal,825ft2is Time (h)

Time (d)

[Ca] (mg/L)

Ca Release (mg/cm2) 0 96 192 288 384 480 576 672 0

2 4

6 8

10 12 14 16 18 20 22 24 26 28 30 0

20 40 60 80 100 120 140 160 180 200 0

8 16 24 32 40 48 56 64 pH 7 pH 8

Page11of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

considereduncoatedduetothebreakjetimpactingcoatedwalls.Theremaining500ft2is marginfordamagedconcretecoatingincontainment.Nocalciumsilicateinsulationexists withinthelooproomsandcalciumsilicateinsulationisnotapartofanydebrisloadsincethe smallamountofcalciumsilicateinsulationoutsidethelooproomsissteeljacketedandnot subjecttodissolution.Calciumreleasesduetodegradationofotherdislodgedinsulationare includedinthetotalcalciumreleaseusedinthetesting.Basedontheconservativeestimatesof existinguncoatedconcrete,therewillbesignificantlylesscalciumreleasedintothecontainment sumpwaterthanwastested.

2. Inthebenchtoptesting,TSPinhibitedcalciumreleasefromuncoatedconcrete.Identicaltests wereruninthebenchtoptestingtodeterminetheeffectofTSPoncalciumconcentration.

Bothsetsoftestswereconductedwithscaledamountsofconcreteandfibrousinsulation.In onesetoftests,noTSPwasused.Inanidenticalsetoftests,arepresentativeconcentrationof TSPwasestablishedinthetestwater.AtpH7,theexpectedcalciumconcentrationin containmentintheabsenceofTSPis40.4mg/LbasedontestswithoutTSPpresent.Inthe presenceofTSP,the30daycalciumconcentrationinbenchtoptestingwas<10mg/L.Inthe absenceofTSP,theconcretecouponsinthetestshowedsignificantdissolution.Inthetests withTSPpresent,concretecouponsinthetestshowednoevidenceofdissolutionand experiencedlessthana1%lossinmass.Forconservatism,theresultsfromcalciumdissolution testswithoutTSPpresentwereusedtodeterminetheamountofcalciumtoaddtotheRig89 (chemicaleffects)testtank.

3. Concreteusedintestingwasnotsafetyrelatedconcreteandthuswasmorelikelytodegradein thebenchtoptestingthanisthesafetyrelatedconcreteinstalledincontainment.

4. ConcretedissolutiondataforpH7wasusedinthetestingtodeterminetheamountofcalcium releasedandtheamountofcalciumusedinchemicaleffectstesting.ThepHintheMPS2 containmentwaterisexpectedtobeabove8.0followingtheLOCAresultinginmuchless calciumrelease.ConcretedissolutionismuchlowerathigherpHasseenintheanswerto question12above.ExpectedlongtermcalciumconcentrationatpH8(withoutTSP)is23.7 mg/Lascomparedtotheexpected(andtested)calciumconcentrationatpH7(withoutTSP)of 40.4mg/L.Thus,thecalciumconcentrationincontainmentislikelytobeasmuchas40%lower thanthetestedvaluedueonlytothepHincontainment.

5. Atotalof15calciumadditionsweremadetotheMPS2Rig89test.Theseadditionshada minimalimpactonheadlossthoughTSPwaspresentinthetesttankattheexpected concentrationincontainment.ThisTSPconcentrationfarexceededtheamountneededto precipitatealloftheavailablecalciuminthetest.Thefirstcalciumadditionwasmadetogether withanaluminumadditionandtheheadlossincreasedfrom0.26to0.66psig.Thesecond calciumadditionwasmadefollowingapowerlossinthetestrigandthatadditionincreasedthe

Page12of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

headlossfrom0.60to0.67psig.Theremaining13calciumadditions(allmadeseparatelyfrom aluminumadditions)hadnosignificantimpactonheadloss.

Theaboveinformationdemonstratesthatsufficientconservatismexistsinthedeterminationofpost LOCAsumpwatercalciumconcentrationtooffsetthepotentiallowersolubilityofcalciumatthehigher postLOCAsumptemperaturesexpectedearlyintheaccident.

13BResponsetoMPS2,ChemicalEffectsQuestion15 TheWCAP16530basemodelisanempiricalmodelofthealuminumreleaserate(RR)basedonthedata setdescribedbyLaneetal[X8X],whichincludeddatafromICET1,CR6873,WCAP7153Aand WCAP16530.TheWCAPmodelisdescribedbyEquation3andtheresultsareshowninXFigure6X.

Equation3 Figure6:3DIllustrationoftheWCAPAluminumReleaseModel

TheAECLmodelisasemiempiricalmodelofthealuminumreleaserate,inthattheequationformwas developedfromfirstprinciplesbuttheparameterswerefittoliteraturedata.Thereleaseequation takesanArrheniusformwithtemperatureand,sincethecorrosionreactioninvolveshydroxide,the releaserateislikewiserelatedtotheexponentialofthepH.Thedatasetusedtofitthemodelwas describedbyGuzonasandQiu[D9D]andwasverysimilartothatusedfortheWCAP16530model.The AECLmodelisdescribedbyEquation4andtheresultsareshowninXFigure7X.

Page13of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Equation4 Figure7:3DIllustrationoftheAECLAluminumReleaseModel

BothmodelsignoreanytimedependenceoftheAlreleaserate.Asonemightexpect,thetwomodels givesimilarpredictions.Mathematicalcomparisonofthetwomodelsshowsthattheydiffermainlyat temperaturesabovethenormalboilingpointofwater.TheWCAPmodelpredictshigherreleaseat moderatepHvalues(betweenpH79.5)andlowerreleaseathighpHvalues,asshowninXFigure8X.At moremoderatetemperatures,thetwomodelspredictverysimilarreleaserates.Forexample,ICET Test5[X9X]wasconductedat60CatpH8.08.5,andbothmodelsareobservedtoconservativelypredict thelongtermaluminumrelease,especiallywhenreleasefromsprayedaluminumwithhighpHsprayis included(XFigure9X).

Page14of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure8:3DDifferentialofWCAPandAECLAluminumReleaseModels

Figure9:WCAPandAECLAluminumReleaseModelsPredictionsofICETTest5Aluminum Concentration

NoteICETTest5concentrationdataadaptedfrom[X9X].SpraypH,reportedas<12,wastakentobe11 forcalculations.

Time (d)

[Al] (mg/L) 0 3

6 9

12 15 18 21 24 27 30 0

50 100 150 200 250 300 350 Submerged Al Only Submerged and Sprayed Al AECL Model WCAP 16530 Model ICET Test 5

Page15of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

TheMPS2postLOCAsumpandsprayoperatesmainlyintherangeofpH8.08.3,wheretheWCAP modelpredictsagreateraluminumreleaserateathightemperaturesthantheAECLmodel(XFigure8X).

Forthe1876ft2ofsprayedand24ft2ofsubmergedaluminumreportedtobepresentatMPS2[X3X],the WCAPmodelpredicts11.9kgAlwhereastheAECLmodelpredicts5.8kgAl(XFigure10X).Itshouldbe notedthatthescaledequivalentof6.6kgAlwasaddedduringtheRig89testF2 Fandthatthelast5 aluminumadditions(XFigure11X),representingover60%ofthealuminumadded,didnotproduce increasesinheadloss,suggestingaheadlossplateau.

Figure10:ComparisonofAECL/WCAPAluminumReleaseModelsPredictionsofSubmerged, SprayedandTotal(Combined)AluminumReleaseforMPS2PostLOCAContainment

2Althoughthescaledequivalentof6.6kgAlwasaddedduringthetest,only6.52kgcanbesaidtohave precipitatedwithcertainty(i.e.,thealuminumloadonthestrainer),asitmustbeconservativelyassumedthat thealuminumconcentrationisnotzerobutthemethoddetectionlimitforICPOESforaluminum(0.4mg/L).

Time (d)

Al Released (kg) 0 2

4 6

8 10 12 0

5 10 15 20 25 30 0

5 10 15 20 25 30 Spray ceases after 40 h WCAP AECL Submerged Sprayed Total

Page16of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure11:Rig89HeadLossTraceCorrectedtoMatchtheApproachVelocityofMPS2

Without30dayaluminumcorrosiontestswheretemperatures(andpressures)oftheMPS2sumpare simulated,itisdifficulttospeculateonthesignificanceofthedifferencebetweenpredictionsofthe WCAPandAECLmodels.TheonlyavailabledataforaluminumreleaseatpH8fortemperatures exceedingthenormalboilingpointofwaterwasreportedfora90minutetestat265F(129C)byLane etal[X8X];thereportedreleaserateof6.6mg/(m2s)wasmanytimesgreaterthanthatpredictedby eithermodel(theWCAPmodelpredicts2.7mg/(m2s),andtheAECLmodelpredicts1.0mg/(m2s)).

Whilethiscomparisonmayseemtohighlightapparentdeficienciesinbothmodels,thedeficienciesof thedatasetaremoreapparent,asitcannotbesaidwithanycertaintythatthevalueof6.6mg/(m2s)is eitheraccurateorrepeatable.Therearemanyvariablestocontrolincorrosiontests,anditisdifficultto getconsistentresults;hence,Laneetal[X8X]couldmeasureareleaserateof0.75mg/(m2s)atpH8and 190F(88C)whileotherscouldmeasurelowerratesatmoresevereconditions:Reidetal[D10D]

measured0.13mg/(m2s)atpH8and200F(93C),Belletal[D11D]measured0.20mg/(m2s)atpH8and 210F(99C),Jainetal[D12D]measured0.53mg/(m2s)atpH10and194F(90C).Thesevaluesare comparedtoWCAPandAECLmodelpredictionsatpH8inXFigure12X.Itisclearthereisalargescatterin thetestdata,withtwodatapointsclusteredcloselytogetherandoneverymuchhigher.Thismay reflectdifferencesintestmethodologyorconditions;AECLhasfoundexperimentaluncertaintiesof about30%innominallyidenticaltests.Bothmodelspredictreleaserateswithinthescatterofthe plotteddata;theAECLmodelbetterfitsmostofthedata,buttheWCAPmodelmorecloselymodelsthe 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10-May 17-May 24-May 31-May 07-Jun 14-Jun 21-Jun 28-Jun 05-Jul 12-Jul 19-Jul 26-Jul 02-Aug 09-Aug Corrected Head Loss (psi) 1 2

3 4

1 2

3 4

Stoppage Events Pump stopped to change a check valve.

Pump stopped unexpectedly.

Power loss.

Power loss.

1st Al Addition 1st Ca Addition 2nd Al Addition 3

4 5 6 7 8

Al Additions 3

4 5

6 7

8 9

10 Ca Additions 11 12 13 14 15 Ca Additions 2

Page17of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

averagevalueandisthemoreconservative.However,thelimitedexperimentaldataavailabledonot provideabasisforselectingonemodelovertheother,andnosignificancecanbeascribedtothe differencesinthepredictedaluminumrelease.

Figure12:ComparisonofAECLandWCAPAluminumReleaseModelPredictionsand MeasuredValuesatpH8

Itshouldalsobenotedthatneithermodelwasdevelopedtopredictshorttermreleaserates.Although shorttermreleaseratesmaybehigherthanpredictedbythemodels,longtermreleaseratesarelikely tobelowerthanpredicted,asindicatedbytheresultsofICETTest5(XFigure9X)andothertestsshowinga plateauinreleaserates,includingtheclassicaluminumcorrosiontestsdescribedbyTroutner[D13D,D14D].

3BMillstonePowerStationUnit3(MPS3),HeadLossandVortexing,RAI6 PleaseprovidethefollowingadditionalinformationtodocumentthattheMPS3strainerevaluation providesadequateassurancethatitwillperformasrequiredunderaccidentconditions:

1. TheDecember18,2008,DNCletterprovidescontradictoryinformationontheamountoffibrous debrisaddedduringthetest.Onpage8,Attachment2,itisstatedthatthelimitingbedwas determinedtobe1/4inchduringearliertesting.Yetthesameparagraphstatesthatonlytwo increments,containingfibrousdebristoform1/16inchbedeach,wereaddedtothetestand thatnofurtherfiberwasadded.Page16statesthattwo1/16inchadditionsweremadeand impliesthattwofurtheradditionsweremadelater.Inaddition,thegraphonpage19shows4 Temperature (ºF)

Temperature (ºC)

Al Release Rate (mg/(m2*s))

180 185 190 195 200 205 210 215 220 120 122.5 125 127.5 130 132.5 135 137.5 140 0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Lane et al.

Reid et al.

Bell et al.

WCAP AECL

Page18of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

fibrousadditions.Describe,indetail,theinitialfibrousdebrisconditionsofthetestandthe amountofanyadditionsthatweremadeduringthetest.

2. TheDecember18,2008,DNCletterstatesthatthelimitingthinbedforMPS3is1/4inchas determinedbyprevioustesting.However,theheadlossplotonpage19,Attachment2,indicates thatthethirdandfourth1/16inchfiberadditionshadlittleeffectonheadloss.Pleaseevaluate thethinbedthicknessforMPS3inconsiderationofthesepoints.Also,ifthethinbedfortheRig 89testisdifferentfromthatofotherteststhatwereusedtoprovideRig89testinputs,please provideanevaluationofhowthefinalqualificationtestcouldhavebeenaffectedbytheuseof suchinputs.Thelicensee'sassertionthat55%ofthedebrisattachedtothestrainerfortheRig 89test,and72%and84%attachedtothestrainerforthereducedscaletestshouldalsobe consideredinthisevaluation.

3. Thedifferenceinheadlossbetweenthetwotestmethodsisaboutanorderofmagnitude.The differencesinnonchemicalheadlossesbetweenthetwotypesoftestswereattributedto contaminantsfromtheuseofriverwaterandtoairevolutioncausedbynonprototypicallylow submergenceduringthereducedscaletests.Itwasstatedthatparticulateandbiologicalactivity intheriverwateraffectedtheheadlossinthereducedscaletesting.Pleaseprovideadditional detailsonhowtheriverwaterparticulateandbiologicalactivityaffectedtheheadloss.Please addressthefollowingitems:

a. Provideanevaluationofthedegreetowhichtheparticulateandbiologicalgrowthfrom theriverwateraffectedtheresultsofMPS3.ItappearsthattheMPS3testswere affectedtoamuchgreaterdegreethanotherAECLtestsconductedundersimilar conditions.PleasediscussthereasonMPS3wasaffectedtoagreaterdegree.

b. Statewhetheranyfiberonlytestswereconductedusingriverwater.Ifsuchtestswere conducted,providetheheadlossesandotherpertinentconditionsforthosetests.

c. Provideanevaluationofthestrainerheadlossresultingfromtheparticulatethatwas containedintheriverwater.Comparetheexpectedtestresult,whentheparticulate fromtheriverwaterandthetestdebrisparticulatearepresent,withtheresultwhen onlythetestdebrisisconsidered.Providetheassumptionsandthebasesforthe assumptionsusedinthisevaluation.

d. Provideanevaluationofwhetherthereducedscaletesting,whichwasusedasaninput fortheRig89qualificationtesting,providedvalidinputduetothenonprototypical biologicalgrowthandparticulatefromtheriverwater.

4. PleaseprovideadditionaldetailsonhowthepostulatedairevolutionaffectedtheMPS3head losstestsconsideringthefollowing:

a. PleaseprovideanevaluationofhowtheairevolutionphenomenonaffectedtheMPS3 testscomparedtootherAECLtestsconductedundersimilarconditions.Pleaseprovide informationonwhyairevolution,asafactorinheadlosses,wouldonlyoccurforAECL strainers.

b. TheresponsetoRAI4statedthattheairevolutionbegantoaffectheadlossassoonas thefibrousdebriswasaddedtothetestandthattheheadlossbegantodecreaseas soonasfibrousdebrisadditionswerestopped.Pleaseprovideanevaluationofwhythe

Page19of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

airevolutionwouldbegintoaffectheadlossasfiberwasaddedtothetestandwhyit wouldstopassoonasfibrousdebrisadditionswerestopped.

c. Pleaseprovideanevaluationofwhytheevolutionofair,causedbytheadditionof fibrousdebriswithairentrainedinit,wouldresultinthehighestheadlosswhena relativelysmallamountoffibrousdebriswasadded.

5. Figure04onpage22,Attachment2,oftheDecember18,2008,lettershowedthatfollowing chemicaldebrisadditionsheadlosswouldincrease,thendecreasebacktothepreaddition value.Pleaseevaluatethisbehaviorconsideringthatitmayhavebeencausedbybed degradation.Considerwhetherhigherheadlossesmayhaveoccurredhadadditionalfibrous debrisbeenpresenttoprovidestructuralsupporttothedebrisbed.

6. PleaseprovideanevaluationofthepotentialforthelowerheadlossintheRig89testing(versus reducedscaletesting)tohavebeencausedbyagglomerationofdebris,especiallyfibrousdebris.

7. PleaseprovideinformationthatjustifiesthatairevolutionwillnotaffectpumpNPSHmarginsor strainerheadlossintheplant.Providethekeyassumptionsusedintheevaluationandthebases fortheseassumptions.

MPS3RAI6AdditionalInformation RAI6(NRCletterdatedDecember17,2008,ADAMSML083230469)originallyaskedforacomparisonof thedifferencebetweenthenonchemicalheadlossesseenintheRig33testingandtheRig89testing andajustificationforthefinalchemicallyladenheadlossusedinthestrainerevaluation.The differencesindebrisonlyheadlosstestingresultsforthetwodifferenttestrigs(Rig33andRig89)were evaluatedduringtheNorthAnnaChemicalEffectsAuditperformedbytheNRCstaffin2008(Reference NorthAnnaPowerStationAuditReportdatedFebruary10,2009,ADAMSML090410626).TheNRCstaff ultimatelyconcludedthat,althoughthereasonsfordifferencesinheadlossforthetwotestrigscould notbedefinitivelyidentified,thesignificantconservatismsincorporatedintothesumpstrainer performanceanalysisboundtheuncertaintiesassociatedwiththedifferenttestresults.Asimilarcase canbemadefortheMPS3sumpstrainerperformance.Adiscussionoftheconservatismsassociated withtheMPS3sumpstrainerperformanceanalysisisprovidedbelow.

Section1.CofAttachment2oftheMillstoneupdatedsupplementalresponsedatedDecember18,2008 (ADAMSML083650005)describestheextensiveplantconservatismsassociatedwiththedesignofthe MPS3containmentsumpstrainer.AdditionalconservatismsarediscussedinSection1.Cofthe MillstonesupplementalresponsedatedFebruary29,2008(ADAMSML080650561).Theoverall magnitudeofeachconservatismcannotbequantified.Howevertheyareviewedtobesignificantand consideringthefactorslistedbelow,whenappliedinacumulativenature,provideassurancetheoverall headlossresultsareconservative.

10%marginwasaddedtothecoatingsparticulatedebrisquantitiesgeneratedfromthezoneof influence(ZOI)andfromunqualifiedcoatings(atotalof2.1ft3ofcoatingsmargin).Reductionof

Page20of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

coatingdebris,whichisallmodeledasparticulate,wouldresultinareductioninthinbedhead loss.

Allunqualifiedcoatingwasdeemedtofailimmediatelyastransportableparticulate.Thisis particularlyconservativesinceunqualifiedcoatingmakesup45%ofthetotaltestedcoatingload and34%ofthetotalparticulateloadonthestrainer.ElectricPowerResearchInstitute(EPRI) testinghasshownthatlessthanonethirdofunqualifiedcoatingsactuallyfailedwhensubjected todesignbasisaccident(DBA)testing.

5%marginwasaddedtothefibrousdebrisquantitiesgeneratedfromtheZOI(atotalofover60 ft3offibermargin).

5%marginwasaddedtothemicrothermdebrisquantitygeneratedfromtheZOI(atotalof0.1 ft3ofmicrothermmargin).

InbothRig33andRig89testing,allfibrousdebriswasconservativelypreparedassinglefine.

100%debristransportwasassumedforcoatings,microtherm,andlatentdebris.

Asacrificialstrainerareaof655ft2wasinstalledforMPS3.

Theeffectiveinstalledstrainerarea(4544ft2)exceedsthetestedstrainerarea(4290ft2).The effectiveinstalledstrainerareadoesnotincludethe655ft2ofsacrificialareawhichisalso installedincontainment.Totalstrainerareainstalledisnearly5200ft2.

DebrisloadrefinementsaftertheRig33test(andbeforetheRig89test)ledtoareductionof about10%intotalparticulatewhichwouldleadtoareductionofthinbedheadloss.

SumpstrainerreducedscalethinbedtestswereinitiallyconductedinRig33todeterminethetotal strainersurfacearearequiredforMPS3.Rig89wasusedtoinvestigatetheinfluenceofchemical precipitatesonthedebrisbedheadloss.Thestrainersupplier,AECL,haspreparedadetailedanalysis reporttoevaluatethedifferentresultsobservedfortheheadlosstestsperformedinRigs33and89.

Theevaluationfocusedonthetestrigconfigurations,flowpatterns,debriscompositionsandquantities, debrispreparation,airbubblegeneration,chemicalenvironment,anddebrisbedformation.AECLand DominionconcludetheRig89testresultsprovideconservativeevidencetoverifytheinstalledstrainer foreachunitwillfunctionundershorttermandlongtermdesignconditions.Rig89testsincorporate lessonslearnedfromtheearlierRig33testing,suchasbiologicalgrowth,testingfluidimpurity,andnon prototypicalstrainersubmergence.Consequently,Rig89providesmoreaccurateresults.The AECL/DominiontestingprogramhasconcludedtheRig89headlosstestresultsareboundingand conservative.

Nevertheless,theevaluationpresentedhereusesthemaximumnonchemicaldebrisbedheadlossfrom Rig33todeterminelongtermmarginsforNPSHandflashing.Thesemarginsarethencomparedtothe headlossfromchemicaleffectstodemonstrateexistingmarginevenwithworstcasedebrisbedhead loss.

XTable3XbelowcomparesthestrainerdebrisheadlosstestresultsfromRigs33and89.Themaximum nonchemicaldebrisbedheadloss(5.1psidat104°F)iscorrectedfortheviscositydifferencebetween thetesttemperatureandtheminimumsaturationtemperatureofcontainment.Thisvalueisconverted tofeetofwater(5.8ftwaterat195°F)foruseinXTable4X,XTable5X,andXTable6X.

Page21of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Forthisevaluation,theimpactofheadlossfromchemicaleffectsisdeterminedbyfindingthe differencebetweentheRig89peakchemicalheadloss(2.2psid)andtheRig89nonchemicaldebris bedheadloss(0.43psid).ChemicalEffectsdebrisbedheadloss(1.8psidat104°F)isconvertedto4.2ft ofwaterwithnocorrectionforlowerviscosityathighertemperatures.

Table3:MPS3Rig33andRig89StrainerTestResults Rig33DebrisHead Loss(psid)at104°F HeadLossdueto DebrisBed(without chemicaleffects)

Rig89Debris HeadLoss (psid)at104°F Rig89PeakHead LosswithChemical Effects(psid)at 104°F HeadLossdueto ChemicalEffects at104°F 5.1(TestM32) 4.6(TestM316) 5.8ftwaterat 195°F 0.43(TestM3 C1) 2.2(TestM3C1) 1.8psid 4.2ftwater

XTable4X,XTable5X,andXTable6XsummarizethekeymarginsavailableforNPSHandflashinginboththe shorttermandlongterm.Longtermmarginssummarizedinthetablesbelowincludethehighestnon chemicaldebrisheadlosstestresult(Rig33,testM32).Eachoftheselongtermmarginsboundsthe headlossduetochemicaleffectsfromtheRig89test.

ForMPS3,thefourRecirculationSpraySystem(RSS)pumpsaretherecirculationpumpsfor containment.TheystartonanRefuelingWaterStorageTank(RWST)levelsignalwhentheRWSTis approximatelyhalffull.Theminimumwaterlevelincontainmentonpumpstartcoversthestrainer.

Quenchspraypumps(whichhavepreviouslystarted)continuetopumptheremainingvolumeofthe RWSTintocontainmentoverapproximately3hoursafterRSSpumpstart.Thusthewaterlevelin containmentiscontinuouslyrisingforapproximately3hoursfollowingRSSpumpstart.Thefinal resultingminimumwaterlevelisapproximately5ftabovethetopofthestrainer.

ForXTable4X,XTable5X,andXTable6XXbelowX,shorttermdataiscalculatedforthetimejustaftertheRSS pumpsstartwhenthecontainmentwaterlevelisatitsminimumvalue,approximatelyhalfoftheRWST (about475,000gallons)remainstobeaddedtocontainment,andthesumpwaterisconsideredtobe saturatedforaLBLOCA.Nodebrisbedisconsideredtobeformed.Thinbedformationtakesatleast6 hoursbasedontestresultsandthewateradditionratefarexceedsworstcasedebrisbedheadlossrate ofchangeasthedebrisbedforms.

Longtermdataiscalculatedforthetimewhichisapproximately3hoursaftertheRSSpumpsstartwhen thenonchemicaldebrisbedisconservativelyconsideredtobefullyformed,aminimumof approximately5ftofadditionalsubmergenceisontopofthestrainerduetoRWSTwateraddition,and aminimumofsumpwatercoolinghasoccurred.Thetemperatureusedforthelongtermcalculationof subcoolingmargin(182°F)isthemaximumtemperatureofthesumpwaterforanyaccidentcasewhen theminimumavailablevolumeoftheRWSThasbeenpumpedtocontainment.Theinitialtemperature forthiscalculationofsubcooling(195°F)isthesaturationtemperaturefortheminimumpressurein containment(10.4psia).

Page22of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Innote3forXTable5XandXTable6XXbelowX,thenonuniformcleanstrainerheadlossisthelossexpectedfor thecleanstrainerwithoutanydebris.Uniformcleanstrainerheadlossisthelossexpectedwhendebris builduphascreateduniformflowthroughoutallofthestrainermodules.Thenonuniformheadlossis lessthantheuniformheadlosssinceflowforthecleanstrainerisexpectedtopreferentiallygothrough thepartofthestrainerwhichisinstalledoverthesumppit.Theotherpartofthestrainersitsovera channelwhichleadstothesumppit.Flowlossesinthischannelarerelativelysignificantandareonly expectedtocontributesignificantheadlosswhenthereisuniformflowacrossadebrisloadedstrainer.

Table4:SummaryofShorttermandLongtermPumpNPSHMarginsatMaximumFlow Case Minimum NPSHavailable (ftH2O)1 MinimumLong TermWaterHeight abovestrainer (ftH2O)2 TotalStrainer HeadLoss (ftH2O)3 NPSH Required (ftH2O)

Minimum NPSHMargin (ftH2O)

RSSPump NPSH(short term) 22.9ft

6.2 4.0 12.7 RSSPump NPSH(long term) 22.9ft 4.9 6.2 4.0 17.6

1) TheminimumNPSHavailableincludestheminimumsumpwaterlevelandaccountsforall pumpsuctionheadlossesexceptforthestrainer.

2) Inthelongterm,theminimumwaterheightabovethestrainerisfromtheremainderofthe RWSTwaterbeingpumpedintocontainment.

3) TotalStrainerheadlossfortheboththeshorttermandlongtermNPSHcalculationsistheclean strainerheadloss(0.4ftwaterat100°F)addedtotherig33debrisonlyheadlosscorrectedfor theminimumsaturationtemperatureduetotheviscositydifference(5.8ftwaterat195°F).

Page23of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Table5:SummaryofShorttermandLongtermStrainerFlashingMarginsatMaximumFlow Case Minimumwater heightavailable (ftH2O)1 MinimumLong TermWaterHeight abovestrainer (ftH2O)2 Margindueto subcoolingto 182°F(ftH2O)

Total Strainer HeadLoss (ftH2O)3 Flashing Margin (ftH2O)

CleanStrainer Flashing(short term) 0.7

0.1 0.6 DebrisLoaded StrainerFlashing (longterm)

4.9 6.1 6.2 4.8

1) Theminimumwaterheightavailableincludestheshorttermminimumsumpwaterlevelfora LBLOCA.ThiscalculationresultsarenotpresentedforaSBLOCAsincethesumpwaterfora SBLOCAissignificantlysubcooled.

2) Inthelongterm,theminimumwaterheightabovethestrainerisfromtheremainderofthe RWSTwaterbeingpumpedintocontainment.

3) TotalStrainerheadlossfortheshorttermflashingcalculationisthecleanstrainerheadlossfor nonuniformflow(0.1ftwaterat100°F).TotalstrainerheadlossforthelongtermisRig33 nonchemicalheadloss(5.1psidat104°F)correctedfortheminimumsaturationtemperature duetotheviscositydifference(5.8ftwaterat195°F)addedtocleanstrainerheadloss(0.4ftat 100°F)foruniformflow.

Table6:SummaryofShorttermandLongtermRSSPumpSuctionLineFlashingMarginsat MaxFlow Case Minimumwater heightavailable (ftH2O)1 MinimumLongTerm WaterHeightabove strainer (ftH2O)2 Margindueto subcoolingto 182°F (ftH2O)

Total Strainer HeadLoss (ftH2O)3 Flashing Margin (ftH2O)

SuctionLine Flashing(short term) 4.5

0.1 4.4 SuctionLine Flashing(long term) 5.7 4.9 6.1 6.2 10.5

1) Theminimumwaterheightavailableincludestheshorttermminimumsumpwaterleveland alsoaccountsforallpumpsuctionheadlossesexceptforthestrainer.Minimumwaterheight availableinthelongtermincludesalowerheadlossfromthesuctionpipingduetolower steadystateRSSpumpflow.

2) Inthelongterm,theminimumwaterheightabovethestrainerisfromtheremainderofthe RWSTwaterbeingpumpedintocontainment.

3) TotalStrainerheadlossfortheshorttermflashingcalculationisthecleanstrainerheadlossfor nonuniformflow(0.1ftwaterat100°F).TotalstrainerheadlossforthelongtermisRig33 nonchemicalheadloss(5.1psidat104°F)correctedfortheminimumsaturationtemperature

Page24of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

duetotheviscositydifference(5.8ftwaterat195°F).Thisisaddedtocleanstrainerheadloss (0.4ftat100°F)foruniformflow.

14BInlightoftheaboveinformation,thefollowingresponsesareprovidedtotheadditionalquestions relatedtoMPS3RAI6posedinNRCletterdatedFebruary4th,2010(ADAMSML100070068).

15BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue1 FortheMPS3Rig89chemicaleffectstestM3C1,fourfibrousdebrisadditionsweremadetothetest looptoachieveathinbedthicknessof1/4inchasdeterminedbypreviousthinbedtests.Thefirstfiber addition(1/16in.(1.6mm))wasmadeat1504h,May30,2008,aftertheadditionoftheparticulate debris.Thesecondfiberaddition(anadditional1/16in.(1.6mm))wasmadeat1750h,May30,2008.

Thethirdandthefourthfiberadditions(each1/16in.)weremadeat0856hand1120h,May31,2008, respectively.Thedetaileddebrisadditioninformationisalsoindicatedintheheadlossvs.timecurveas showninXFigure13X.

Figure13:DebrisBedHeadLossvs.TimeforTestM3C1

16BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue2 AsshowninXFigure13X(responsetoRAI6issue1),thedebrisbedheadlossincreasedfrom0.38psito 0.43psiafterthethirdfiberaddition,indicatingthatthethinbedthicknesswasatleastequivalentto threeadditions.Afterthefourthfiberaddition,theheadlosspeakedat0.45psiandstabilizedat0.43 psibeforethefirstchemicaladdition.Thisindicatesthatthefourthadditionmadelittledifferenceand thethinbedcouldbeconsideredtobelessthan1/4inch.

Theparticulatedebrisloadwas10%lowerinRig89testthaninRig33tests.Thus,thethinbed thicknesswouldbeslightlylowerthanthatofRig33tests,eventhoughittookthesamefourfiber M3-C1 Debris Bed Head Loss 0

4.5 9

13.5 18 22.5 27 31.5 36 40.5 45 30/May/08 12:00 30/May/08 18:00 31/May/08 0:00 31/May/08 6:00 31/May/08 12:00 31/May/08 18:00 01/Jun/08 0:00 Time (standard)

Temperature (ºC), Flowrate (USGPM) 0 0.06 0.12 0.18 0.24 0.3 0.36 0.42 0.48 0.54 0.6 Head Loss (psi)

Temperature (ºC)

Flowrate (USGPM)

Head loss (PSI) 1st fiber addition 2nd fiber addition 3rd fiber addition 4th fiber addition power loss at 1443 h, May 31 power restored at 1535 h, May 31

Page25of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

additionstoformathinbed.Also,inRig33,periodicfloorsweepingandcontinuousstirringwouldhelp maintainfibersuspendedandeventuallyattachedtothedebrisbed.Thesetwofactorswouldcausea higherpercentageofdebristoattachtothestrainersurface.TheRig89testheadlossversustime curveconclusivelyshowedthatathinbedwasformedbythefourthfiberaddition.Extrafiberaddition wouldnotincreasetheheadlossandwouldnotdecreasetheheadlossbecauseanyextrafiberlayson topofthealreadyestablishedthinbed.Thelowerpercentageofdebrisattachedtothestrainersurface ascomparedtothatoftheRig33testshadnonegativeeffectsonthestabilizedheadloss.

17BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue3a OttawaRiverwaterwasusedintheRig33tests,whiledistilledwaterwasusedintheRig89test.

BacteriagrowthandtheresultingbiologicaleffectswereobservedduringRig33testingforSurryin May2006[D15D]andforMPS3inOctober2006[D16D].Fortheaffectedtests,biologicalactivityprevented headlossfromstabilizingafterthesecondfiberaddition.Slimeformingwasbelievedtobethemajor mechanismforbiologicaleffects.Riverwaterparticlesalsocontributedtothehigherheadlossobserved inthosetests.

AccordingtoAECLreport(AECL1124)[D17D],seasonalslimeformationinsystemsusingwaterfromthe riverhasbeenaproblemsince1946.Slimeformingmicroorganismshavetheabilitytogrowrapidly underfavorableenvironmentalconditions.Theseorganismsmaybebacteria,fungi,algaeormoldsand thefactorseffectingtheirgrowtharetemperature,pH,nutrientsandconcentrationofelectrolytes.All ofthesemicroorganismsrequireasourceofcarbonforgrowth.(IntheRig33tests,walnutshellflour couldbetheidealcarbonsource.Theriverwaterparticulatecouldbeanothercarbonsource.).Also reportedintheAECLreport1124,asampleoftheslimewassenttotheNationalAluminateChemicals Companyformicrobiologicalidentification.Thereportindicatedthatitconsistedmainlyoffungal filamentsandcrystallinematerial.Therewereoccasionalbacteriaanddiatoms(Fragilaria)present.

ThewaterlevelheightoftheOttawaRivercouldalsoaffecttheconcentrationofriverparticlesand slimeformingorganisms.HigherwaterlevelresultedinhigherpeakheadlossasshowninXFigure14X.

Page26of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure14:OttawaRiverWaterLevelHighandPeakMPS3Rig33TestHeadLoss

TwodifferentwatertreatmentmethodswereusedinMPS3Rig33teststoinhibitbiologicaleffects beforedebrisaddition.Thewatertreatmentmethodshowntoeffectivelyinhibitbiologicaleffects observedinSurrytestingconductedinlateMay2006wasusedforTestsM31toM310,whichwere conductedinSeptemberandOctober2006.Nitricacid(5molar)wasaddedtothetestwaterto decreasethewaterpHtoatargetvalueof5.5(rangeof5.1to5.6pH).Thetestwaterwasthenheated totesttemperature.Oncethewatertemperaturewasstable,sodiumhydroxidewasaddedtoincrease thewaterpHtoatargetvalueof6.8(rangeof6.5to7.0pH).Afterdebriswasadded,therewasno furtherbiologicalcontrol.

AmoreaggressivewatertreatmentwasdevelopedinNovember2006followinganapparentrecurrence ofbiologicaleffectsinOctober2006thataffectedTestsM36toM310.Thistreatment,consistingofa combinationofchlorineadditions,heatingtoahigherwatertemperatureandwaterfiltering,wasused forTestsM314andM316.Notethatfilteringwasinstitutedtoreducethequantityofparticulateinthe testwater,nottoinhibitbacterialgrowth.Withthistreatment,sufficientchlorinewasaddedtothetest watertomaintaintheconcentrationabove10ppmduringsubsequentheatingandfiltering,as concentrationsofthismagnitudehavebeenshowntopreventbacterialgrowth.Thewaterheatup procedurewaschangedtoheatthewatertoahighertemperaturethanusedpreviously(136F(58C) versus122F(50C))beforecoolingtothetesttemperature,aswatertemperaturesapproaching140F (60C)aresufficienttokillmanytypesofbacteria.Bagtypefilterslocatedonthedischargesideofthe Ottawa River Water Level Height and Peak Test Pressure Vs. Date 110.4 110.6 110.8 111 111.2 111.4 111.6 111.8 112 112.2 1-Sep 5-Sep 9-Sep 13-Sep 17-Sep 21-Sep 25-Sep 29-Sep 3-Oct 7-Oct 11-Oct 15-Oct 19-Oct 23-Oct 27-Oct 31-Oct 4-Nov Date Water Level (m) 0 1

2 3

4 5

6 7

8 9

10 Head Loss (Psi)

Water Level Height Head Loss M3HeadLoss (2006)

Page27of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

pumpwereusedtoreducethequantityofparticulateinthewater.(Thisparticulateconsistedofsmall quantitiesofsiltandrustintheservicewaterandresidualwalnutshellflourfromthetestsection and/orpipingsystem.)Twostagefilteringwasemployed:a200mporesizebagfilterwasusedforthe first10hofheatup,anda10mporesizefilterwasusedforthesecond10h.Chlorinewasnotadded tothetesttankafterthefirstdebrisaddition.ThreesamplesofAECLsservicewaterwerecollected duringthetestprogramandanalyzedforTotalSuspendedSolids(TSS).ThelevelsofTSSareshownin XTable7XXbelowX.

Table7:TotalSuspendedSolidsinServiceWater Date PointinTestProgram TSS(mg/L)

Standard Fine*

September1,2006 PriortoProgram 0.2 n/a October13,2006 PriortoTestM38 0.6 n/a November1,2006 PriortoTestM314 1.2 3.0

• NotefineTSSmeasurementsnotmadeforsamplestakenpriortoNovember1,2006.

StandardTSSismeasuredbydrawingthewatersamplethrougha1.5mporeMisafilter.ThefineTSS reportedhereinwasmeasuredbydrawingthewatersamplethroughaspecial0.1mporefilter.

Samplesofthedebrisbedattheendofeachtestwereanalyzedforbiologicalactivity.Thisanalysisis doneusingBiologicalActivityReactionTests(BART)forslimeforming(SLYM)andheterotrophicaerobic bacteria(HAB),followedbymicrobialgrowthonanagarmediaandcellcounts.Analysisresultsare showninXTable8X.BARTresultsareshownaspositive(+)ornegative()formicrobialgrowth.Cell countsareshownascolonyformingunitspermLofwater(CFU/mL).

Table8:BiologicalActivityAnalysisResults Test Sample SLYM HAB CFU/ml M32

1. 1

+

+

4x106

1. 2

+

+

3.7x107 M316

1. 1

+

+

2x107

1. 2

+

+

2x107 NoteSLYM=slimeforming,HAB=heterotrophicaerobic,positive(+)ornegative()formicrobial growth,andCFU/ml=colonyformingunitspermlofwater.

TheanalysisresultsshowthatbacteriawerepresentinthedebrisbedattheendofTestsM32and M316.Therefore,bothwatertreatmentsdidnotentirelyeliminatebiologicaleffects(thetreatment methodmightnotbeeffectiveforfungiand/oralgae).Itwaspostulatedthatbothtreatmentsinhibit thedevelopmentofbiologicaleffectslongenoughtoallowatesttobecompleted,withtheaggressive treatmentprovidingmoretimeand/orbeingmoreeffective.

UsingthecellcountresultsinXTable8X,thetotalcolonyformingunitsintheRig33testM316testwater wouldbe1x1014((2x107/mL)x5000L),whichis5timesgreaterthanthenumberofwalnutshell particles(walnutshellparticles:2.0x1013).Thetesttankvolumeis5000L.Averagebacterialcellis3to 5mindiameter.Inthetestrig,bacteriagrowthaffectingstrainerfunctionwouldformabiofilmon

Page28of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

surfacesthatmaybeonetoafewhundredmicronsthick.Itisassumedthateachcolonyformingunit originatedfromonebacterialcell.Theeffectsofthecolonyformingunitonthedebrisbedheadloss wouldbesignificant.Assumingallthesecolonyformingunitswereseparatesphericalparticles.The massofeachparticlewascalculatedtobe3.3x1011g.Thetotalmassoftheslimeparticleswouldbe3.3 kg(7.3lbm).Inordertoquantifytheheadlossinfluencefromtheslimeparticles,theNUREG/CR6224 correlationwasused.Thecalculationshowsthattheextraheadlossincreasefromtheslimeparticles couldbeashighas1.2psi.

Fortheriverwaterparticulateinfluence,ananalogouscomparisonwasperformedasfollows.Theactual massofsuspendedsolidswascalculatedtobeapproximately0.033lb(3mg/Lx5000L).Assumingthe increaseonheadlossfromtheriverwaterparticlesweresimilartothatofMicrothermandthehead lossinfluencewasproportionaltotheirmass,theheadlossinRig33testcouldbe0.06psihigher(Head lossimpactofriverparticulate4.3psix0.033lb/2.39lb0.06psi).Theheadlossincreasedueto MicrothermadditionwasdemonstratedinTestNA2[D18D],where2.39lbofMicrothermwasaddedto thetestafterabedwasformed.Theheadlossincreasedimmediatelyby4.3psi,toavaluesixtimes greaterthanpriortotheMicrothermaddition,andthetestwasabortedbeforeheadlosshadreacheda stablevalue.HoweveritisnotclearthatriverwaterparticulateandMicrothermhaveequivalentimpact ondebrisbedheadloss.Inanycase,0.06psiisaninsignificantheadlossimpact.

ThereasonthattheMPS3testswereaffectedtoamuchgreaterdegreethanotherRig33testswas becauseofthetestenvironmentsandtheairevolutioninfluence.Testenvironmentsincludethe amountofwalnutshellflourandslimeformingorganismsinthetestwater.Walnutshellflourcould providecarbontoslimeformingorganismsasmentionedbeforeandslimeformingorganisms concentrationsinOttawaRiverwaterfluctuatedseasonallyandwereaffectedbythewaterlevelheight.

Thewaterlevelheightchangedoccasionallyduetoprecipitationand/ordischargefromtheupstream hydrodam.MPS3testshadthehighestwalnutshellflourloadperunitstrainersurfaceareaamongall theDominiontests.Asthedebrisbedheadlossexceededathresholdvalue,inthiscase,thestatichead ofwaterabovethefin,airevolutionoccurred.IntheMPS3Rig33tests,airevolutionwasthedominant factorthatcontributedtothehigherheadlossascomparedtootherRig33tests.

Insummary,severalfactorscollectivelycontributedtothenonchemicalheadlossdifferencesbetween theRig89testandtheRig33testforMPS3.Thesefactorsinclude:

LessparticulatedebrisinRig89testthaninRig33testduetoarefinementofpostLOCAdebris loadcalculation(10%less),

DistilledwaterwasusedintheRig89test,whileOttawaRiverwaterwasusedinRig33test, BiologicalgrowthinRig33testduetotheuseofOttawaRiverwater,whilenobiologicalactivity inRig89test, DebrisusedinRig89wasautoclavedtoeliminatebiologicalgrowthinRig89.Rig33testsdid notuseautoclaveddebris, DebriswasconservativelymaintainedinsuspensioninRig33inaturbulentflowoutsidethetest section.Theturbulentflowwascausedbycontinuousstirringandreturnflowflushing,and, LargeamountofairevolutioninRig33testwhilenoairevolutioninRig89test.

Page29of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

18BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue3b NofiberonlytestwasperformedforMPS3,butaseriesoffiberonlybypasstestswereperformedfor MPS2,NorthAnnaandSurry.Fiberbypasstestswereconductedtodeterminethequantityand characteristicsoffibrousdebristhatpassesthroughthestrainer.Thefullfibrousdebrisloadwasused forthesetests.Noparticulatedebriswasused.Thefibrousdebriswaswashedtoremovedirtanddust fromthefibers.Fibrousdebrisloadwasaddedtothetesttankatthestartofthetestwithin30minutes.

Foreachfiberbypasstest,thesametestpreparationwasfollowedasitscorrespondingthinbedandfull debrisloadtestsintermsoftestwater,heating,watertreatmentanddebrispreparation.Thefiber bypasstestswereusuallyrunforseveralhoursbecausetheheadlossstabilizedveryquickly.Thehighest observedheadlossoccurredinTestM228.Theheadlossstabilizedat0.1psi.Nowatertreatmentor pretestwaterfilteringwasusedforTestM228.ForNorthAnnaandSurryfiberbypasstests,thehead losswasnegligible,forexample,inTestS242(SurryRSfiberbypasstest),theheadlossstabilizedat 0.034psi.ForTestNA19(RSfiberbypasstest)theheadlossstabilizedat0.02psi.Watertreatmentand pretestwaterfilteringwereusedforbothNorthAnnaandSurryfiberbypasstest.Thehighheadloss observedinTestM228mightverifythatparticulatefromOttawaRiverwaterandbiologicalgrowth affecteddebrisbedheadloss.

EventhoughsomeheadlosseffectswereobservedinTestM228,thephenomenonwasnot representativebecausethefiberonlydebrisbedwastooporoustocatchtheminuteriverparticlesand themicroscopicslimeformingorganisms.Riverwaterparticlesandslimeformingorganismscouldcause ahigherheadlossifamorecompactthinbedwasformed.

19BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue3c Notestresultsexistwhichdirectlyexaminetheeffectoftheriverwaterparticulateintheabsenceof othervariables.AsbrieflymentionedinresponsetoRAI6Issue3a,theriverwaterparticlescould increasetheheadlossby0.06psi.Theevaluationwasbasedontheassumptionthattheminuteriver waterparticlewouldbehavethesameasthatoftheMicrothermparticlesonthedebrisbedheadloss.

ThecalculatedriverwaterparticulatemassforTestM316islistedinXTable9X.Themassofriver particulate(0.03lb)isinsignificantandnotexpectedtocauseameasurableimpactonheadloss.

Table9:NumberofParticles Test TSS(fine)[mg/L]

TestWaterVolume[L]

TotalMassofRiverWaterParticulate[lb]

M316 3.0 5,000 3.3x102 20BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue3d Theimpactsofriverwaterparticulateandbiologicaleffectsontherig33headlossresultsarerelatively small.

TheinputsthatweretakenfromtheRig33testsweredebrispreparationandadditionmethodforthin bedformingandthespecificthinbedthickness.Thedebrispreparationandadditionsequencewere acceptedasconservative.

Page30of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

AsshownintheadditionalinformationforRAI6above,useoftheRig33testresultsforthemaximum nonchemicalheadlossleavesadequatemarginforpumpNPSH,strainerflashing,andsuctionline flashingtoboundtheconservativeestimateofheadlossduetochemicaleffects.

21BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue4a Airsolubilityinwaterisproportionaltotheabsolutepressureatthelocationofinterest.Themaximum quantityofairthatcouldbedissolvedinthewaterisproportionaltotheabsolutepressureabovethe watersurface.Instrainertesting,asthedebrisbedheadlossbecomesgreaterthanthestaticheadof waterabovethefin,dissolvedairwillevolvefromthesolution.InMPS3Rig33tests,thewater submergencewassetto8inches.Thecorrespondingstaticheadwas0.29psiatthetopofthe submergedfinand1.4psiatthebottomofa30inchhighfin.Oncethedebrisbedheadlossexceeded 0.29psi,airevolutionwouldstarttooccuralongthetopsofthefinsandairbubbleswouldstartto accumulatewithinthedebrisbed.Whendebrisbedheadlossexceeded1.4psi,airevolutionwould occuralongtheentireheightofthefins.Airbubblesretainedwithinthedebrisbedwouldrestrictthe flow,increasingdebrisheadloss.

Basedonadissolvedaircalculation,itwasfoundthatairevolutionincreasessignificantlyasthehead lossacrossthedebrisbedincreases.XFigure15Xshowsatheoreticalplotofhowairevolutionisaffected bydebrishead.Thisplotassumesthatairwillevolveoutofsolutionimmediatelyitexceedsthe saturationconcentration,whereasitislikelythatthereissometimedelayandtheactualairrelease wouldbelessthanindicated.Nevertheless,forarelativelysmallheadloss,airevolutionisverylowand itstartstobecomemoresignificantasheadlossexceedsapproximately2psi.Thetestsperformedfor otherDominionplantshadheadlosseslessthan2psi,whichwasnotenoughtocausesignificantair evolution.

Ifalltheairthatevolvedfromsolutionweretoremainwithinthedebrisbed,withasignificantheadloss itwouldtakeonlyminutesbeforethedebrisbedwascompletelyblockedbyair.Ofcourse,thereisa constantmigrationofsuchairbubblesthroughthedebrisbed,sothebedwouldnevergetcompletely blocked.Understeadystateconditionsthereisequilibriumbetweenairevolutionwithinthebedandair migrationthroughthebed.

Typically,afterafiberadditiontheheadlossincreasedrapidlytoamuchhighervalue,thenafterawhile theheadlosswoulddropandstabilizetoalowervalue.Theairbubblescaughtinthedebrisbedcan explainthisscenario.Itwasobservedthatairbubblesbecameattachedtothefibersduringthedebris preparationprocessandwereaddedtothetesttankalongwiththedebris,asshowninXFigure19X.

Shortlyafterafiberaddition,theseairattachedbubblesstartedtorestricttheflowpath,whichinitiated theriseinheadlossandthenresultedingenerationofmoreairinsidethedebrisbedduetolow submergence.Eventually,therateofairgenerationbecameequaltotherateofairmigration,andthe headlossstabilizedatalowervaluethanthepeakvalue.Theairbubbleblockageinthedebrisbedis believedtobethemostsignificantfactorforhighheadlossinMPS3testsinRig33.Sincethis mechanismisstrictlydependentonthewatersubmergenceandheadloss,itisexpectedtooccurfor anystrainerdesignundersimilarconditions.

Page31of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure15:TheoreticalAirEvolutionataPointontheStrainerSubmergedby26Inches

Alesssignificantcontributingfactortostrainerheadlossduetoairevolutionisaccumulationofair withinthestrainer.Becauseofthetestmoduleconfiguration,thistendedtooccurinmanyoftheRig33 tests.

AnequationwasdevelopedinReference[D19D]tocalculateheadlossacrossastrainerthatispartiallyair filled:

1/2

Equation5

Where,

=pressuredropacrossthestrainerwiththesameuniformdebrisbedandflow ratewhenthestrainerisfilledwithair,

=pressuredropacrossthesamedebrisbedwhenthestrainerisfilledwithwater, h=heightoftheairvoidwithinthestrainer.

Theextrapressurelossduesolelytothepresenceofairwithinthestrainerisquantifiedbythelast term.Thiseffectisduetothereductionofdrivingpressureforflowtopassthroughtheupperportionof thestrainerascomparedtothelowerportionofthestrainer;thustheupperportionofthestrainer loseseffectiveness.

0.00%

0.10%

0.20%

0.30%

0.40%

0.50%

0.60%

0.70%

0.80%

0 1

2 3

4 5

6 Voidfractionbyvolume DebrisHeadLoss(psi)

Page32of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

ThesecondcolumnfromtherightinXTable10Xquantifiestheextraheadlosscausedsolelybyair accumulationwithintheteststrainerforallDominionRig33tests.ForMPS3,thiscausedanadditional 0.5psiheadloss.ModerateairaccumulationwasalsoobservedduringNorthAnnatests,whichcaused approximately0.3psiextraheadloss.

ThetwophotosbelowshowairbubblesobservedintheMPS3reducedscaletest(XFigure16X)andlarge scaletest(XFigure17X).

Figure16:AirBubblesObservedEruptingfromFinChannelsatPumpStopinTestM316

Page33of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure17:AirBubblesEmergingfromDischargingHeaderinMPS3LargeScaleTest

AirevolutionwasalsoobservedinotherDominionstrainertestsperformedbyAECL,butnotasmuch.

SincethedebrisbedheadlossoftheseothertestswaslowerthanthatoftheMPS3tests,lessairwould begeneratedwithinthedebrisbed.

SimilarairevolutionwouldoccurforanystrainerunderconditionssimilartotheMPS3tested conditions.

Table10:AirEvolutioninRig33Tests Test Strainer Submergence

&FinHeight

[inches]

StaticHead Top~Bottom ofFins[psi]

PeakDebris BedHeadLoss

[psi]

HeadLoss CausedbyAir insideStrainer

[psi]

SignificantAir Evolution?

NAPSLHSI NA15 7/20 0.25~0.97 1.4 0.36 Minor NA16 7/20 0.25~0.97 1.3 0.36 Minor NAPSRS NA10 27/15 0.97~1.5 2.1 0.27 Minor NA14 27/15 0.97~1.5 1.4 0.27 Minor SurryLHSI S233 7/20 0.25~0.97 0.53 0.06 No S235 7/20 0.25~0.97 0.24 0

No SurryRS S228 27/15 0.97~1.5 1.0 0.001 No S230 27/15 0.97~1.5 1.3 0.10 Minor SPS 12/15 0.43~0.97 0.39 0

No

Page34of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Rig33C1 MPS2 M222 7/37.75 0.25~1.6 0.81 0.11 Minor M227 7/37.75 0.25~1.6 0.68 0.07 Minor MPS3 M32 8/30.38 0.29~1.4 5.1 0.54 Major M316 8/30.38 0.29~1.4 3.6 0.54 Major 2BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue4b TheabovementionedRAI4(NRCRequestforAdditionalInformationdatedDecember17,2008)is quotedasbelow:

Theexplanationforhigherpeakheadlossthatoccurredduringlargescalestrainerperformance testingstatedthatairwasreleasedfromsolutionwhenheadlossacrossthedebrisbedlowered thepressureinthedebrisbedbelowthestaticpressureofwaterontopofthedebrisbed.Thisair releaseapparentlyresultsinhigherpeaksinheadloss.Theexplanationofthisphenomenonis unclear.Itisalsounclearastowhythisphenomenonwouldnotoccurduringthereducedscale testingsincetheheadlossesandsubmergenceweresimilar.Pleaseprovideadditionaldetails andevaluationofthecauseofthepeakheadlossthatoccurredduringthistesting.

TheMPS3largescaletestM3L2wasperformedintheAECLlargescalestrainertestingfacilityRig85.

Inthattest,manyairbubbleswereobservedcomingoutofthedischargeheaderafterthethirdfiber addition,asshowninXFigure17X.Thedischargeheaderwaslocatedonthefloorofthetesttank.During thetest,theheadlossstabilizedat2.7psiafterthesecondfiberaddition.Thethirdfiberaddition increasedtheheadlossto4.1psi.Threemorefiberadditionswereaddedintothetestandeach additioncausedaspikeinheadlossasshowninXFigure18X.DNCsresponsetoRAI4,datedMarch13, 2009,referredtothefourth,fifthandsixfiberadditions.Priortothesefiberadditions,airevolutionhad alreadyreachedasignificantlevelduetohighdebrisbedheadloss.

Theobservedheadlossspikeafterthefourth,fifthandsixthfiberadditionwasduetotheairbubbles trappedinsidethefibrousdebris.Microscopicexaminationoffiberspreparedinasimilarfashion(i.e.,

usingapressurewashertoagitateandbreakuptheclumpsoffiber)showedthatairbubbleswere attachedtothefibers(seeXFigure19X).Itwastheairbubblesthatinitiatedthepressurespikes,notthe fibers.

Assoonasthefibersreachedthedebrisbed,thebubblesstartedtomigrateintothedebrisbed, blockingflowareaandcausingtheheadlosstoincrease.Theincreasingheadlosscausedthegeneration ofmorebubbleswithinthebed,which,inturn,causedafurtherincreaseinheadloss.Onceadebris additionwascompletedandnonewbubbleswerearrivingatthedebrisbed,thenthecontinuing migrationofairbubblesthroughthedebrisbedintothefinsbegintodecrease,unblockingflowarea andcausingfurtherheadlossdecreases.Eventually,therateofairgenerationdecreasedtobecome equaltotherateofairmigration,andtheheadlossstabilizedatalowervaluethanthepeakvalue.

Page35of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure18:HeadLossesvs.DebrisAdditioninMPS3LargeScaleThinBedTest

Figure19:AirBubblesAttachedtoPreparedThermalWrapFiber

23BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue4c AsexplainedinresponsetoIssue4b,priortothelastthreefiberadditions,airevolutionalreadyexisted inthesystemduetohighdebrisbedheadloss(4.1psi).Newlyaddedfiberwouldbringentrainedair bubblesintothedebrisbed,blockingflowareaandcausingtheheadlosstoincrease.Theincreasing headlosscausedthegenerationofmorebubbleswithinthebed,which,inturn,causedafurther increaseinheadloss.Onceadebrisadditionwascompletedandnonewbubbleswerearrivingatthe M3L-2 20 24 28 32 36 40 44 48 52 56 60 30/Jan/0 7 4:48 30/Jan/0 7 12:00 30/Jan/0 7 19:12 31/Jan/0 7 2:24 31/Jan/0 7 9:36 31/Jan/0 7 16:48 01/Feb/0 7 0:00 01/Feb/0 7 7:12 01/Feb/0 7 14:24 01/Feb/0 7 21:36 02/Feb/0 7 4:48 02/Feb/0 7 12:00 02/Feb/0 7 19:12 Time Flow rate (L/s); temperature (C) 0 6

12 18 24 30 36 42 48 54 60 Head loss (kPa)

Rig 85 FT-1 Rig 85 TE-2 Rig 85 PDT-1 Rig 85 PDT-2 Rig 85 PDT-3 particulate addition

@0954 1st fiber addition

@1130 2nd fiber addition

@1305 3rd fiber addition

@0853 brush tank floor

@1156 4th fiber addition

@1502 5th fiber addition

@1325 6th fiber addition

@0914 floating debris addition @0853

Page36of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

debrisbed,thenthecontinuingmigrationofairbubblesthroughthedebrisbedintothefinsbeginto decrease,unblockingflowareaandcausingfurtherheadlossdecreases.Eventually,therateofair generationdecreasedtobecomeequaltotherateofairmigration,andtheheadlossstabilizedata lowervaluethanthepeakvalue.

24BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue5 TheheadlossbehaviorafterthechemicaldebrisadditionswasexplainedinAECLtestreportMIL3 34325TR004Rev1[D20D]as Aluminumadditionsinvariablyresultedinheadlosspeaks,followedquicklybydecreasesinhead loss.Thisphenomenonseemstohavebeentheresultoftheadditionmethodandmayhavebeen causedbythetransientlyhigh(andnonprototypical)concentrationofdissolvedaluminum.As thealuminumprecipitatesformedandsettled,theheadlossreturnedtolowervalues.

TheheadlossversustimecurveshowninXFigure13Xdemonstratedthatthethinbedthicknesswas1/4inch orless.Aslongasathinbedwasformed,furtherfiberadditionwouldnotincreasethenonchemical debrisbedheadloss.Extrafiberwouldeitherlooselyattachtothestrainersurfaceformingaporous layer,orsettleonthetankfloor.Aflowsweepattheendofthetestdemonstratedthattheheadloss respondedquicklytochangesinflowrateandheadlosschangeswerefoundtobereversible.Posttest examination(asshowninXFigure20X)alsoconfirmedthatthedebrisbedwasnotdegradedduringthe testandheadlosswasnotlimitedbyholesin,ordislocationof,portionsofthedebrisbed.

Figure20:APieceofDebrisBedaftertheRig89Testing

25BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue6 ThepotentialforthelowerheadlossintheRig89testingtohavebeencausedbyagglomerationof fibrousdebriswasverylow.Fibrousdebriswassprayedassinglefinebyusingahighpressurejetflow ina200Lplasticbarrel.Thesprayedfiberwasthenaddedintotheinlinedebrisadditiontank.The debrisadditiontankwasequippedwithastirrer.Afterabatchoffibrousdebriswasaddedintothetank, thestirrerwasturnedontosuspendthefiberandtoavoiddebrissettlingoragglomeration.Thedebris additiontankwasthenvalvedintoletthefiberflowtothestrainerboxslowlybyadjustmentofthein

Page37of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

linecontrolvalves.XFigure21X,XFigure22XandXFigure23Xshowthatafterthetest,thedebrisbedwasfirm anduniform.Nofibrousdebrisclumpswereobserved.

ThereasonsforlowernonchemicalheadlossinRig89areunclear,howeversufficientmarginexistsas describedabovetoaccountforthehigherRig33nonchemicalheadlossresultsalongwiththeheadloss duetochemicalprecipitantsfoundinRig89.

Figure21:DebrisBedattheEndofMPS3ChemicalEffectsTest

Page38of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure22:CloseUpofaPieceofDebrisBedRemovedfromtheStrainerSurface

Figure23:DebrisBedThicknessafterMPS3Rig89ChemicalEffectsTest

Page39of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

26BResponsetoMPS3,HeadLossandVortexing,RAI6,Issue7 Assumptions:

TheRWSTwillbeemptiedwithinamaximumof3hoursfromthestartoftheaccident.

Theminimumcontainmentwaterlevelabovethetopofthestraineris4.9feetforaSBLOCAand 5.5feetforanLBLOCA.

Themaximumtemperatureofthecontainmentwaterislessthan185°Fthreehoursafterthe accident.Waterdensityat185°Fis60.46lb/ft³.

Thegenerationofairinthedebrisbedisdependentonthestaticheadofwaterabovethefin;ifthe debrisbedheadlossislessthanthestaticheadofwater,noairevolutionisexpected.Thesubmerged depthforthereducedscaletestswassetat8incheswhereastheminimumwaterlevelinMPS3 containmentcontinuestorisefor3hoursfollowingtheaccidenttoaminimumheightof4.9ftabove thetopofthestrainer.

Boththemaximumstaticheadandtheincreaseinstaticheadwithtimeincontainmentmustbe comparedtotheheadlossresultstodetermineifaircouldbegeneratedinthedebrisbedinMPS3 containment.

Atminimumsubmergence(4.9ft),thestaticheadatthetopsofthefinsincontainmentwillbe2.0psi with185°Fwater.Themaximumdebrisbedpressuredropwasseenas2.17psiinRig89testing.This maximumheadlossvaluewaslatercalculatedtobe1.67psiduetothesizeofthetestmodulebeing 5.08ft²vicethe5.74ft²thoughttobethesurfaceareaatthetimeoftesting.Themaximumdebrisbed headlossisduetoadditionofaluminumprecipitatestothedebrisbed(fromaluminumcorrosion).

Aluminumcorrosionisalongtermphenomenonwhichwillonlyaddparticulatetothedebrisbedlong afterthewaterhascooledresultinginsignificantadditionalstaticheadduetosubcooling.Thus,the debrisbedheadlosswillremainbelowthestaticheadonthestrainerpreventingairevolutioninthe debrisbedorstrainer.

Withinthefirstthreehoursaftertheaccident,thestaticwaterheadlosswouldincreasetoatleast2.0 psi.ThesumpwaterturnovertimeatthestartoftheRSSpumpswouldbe57minutes(sumpwater mass:3,819,002lb,flowrate:8220USGPM[D21D]).Itonlytakes3turnoversforthestaticwaterheadto reachatleast2.0psi.TheRig89testturnovertimewas5minutes.InRig89,3turnovers(15minutes) afterthefirstfiberaddition,thedebrisbedheadlossbarelyreached0.1psi.Asobservedinthestrainer testing,itusuallytookdaystobuildathinbed.Thus,airevolutionwillnotoccurintheplantstrainers.

4BMPS3,NetPositiveSuctionHead,RAI9 Itisnotclearhowwaterdrainsfromtherefuelingcavityintothereactorcavity,andwhetherthis drainagepathislargeenoughtoensurethatdebrisblockagewouldnotoccur.WhiletheplantFinal SafetyAnalysisReport(FSAR)documentsthatasignificantamountofventingsurfaceisavailable,there isalsoasignificantquantityofdebrisavailable.Thepotentialforblockageoftheventcoversisalso consideredintheFSAR.

Page40of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

TheRAIintendedtoaskabouttheentirerefuelingcavity:didyourresponseaccountfortheentire refuelingcavityoronlythecavitysaddle?IfyourRAIresponsedidnotaccountfortheentirerefueling cavity,pleaseupdateyourresponse.

Toensurethattheevaluationhasaccountedfortheworstcaseminimumcontainmentwaterlevel, pleaseclarifythedrainagepathfromtherefuelingcavitytothereactorcavity,theminimumflow restrictions,andprovideabasisforwhyblockagewouldnotoccurthere.

27BResponsetoMPS3,NetPositiveSuctionHead,RAI9 ThepreviousRAIresponse(seeAttachment2toDNCLetterSerial09175datedMarch13,2009) consideredthemaximumpotentialholdupvolumeoftherefuelingcavity.Theminimumwaterlevel calculationconservativelydeterminestheminimumcontainmentwaterlevelwhichexistsattheearliest RSSpumpstarttime.Thetotalpossibleholdupintherefuelingcavityislimitedto49,202gallonssince anywaterbeyondthisvolumespillsintothereactorcavityandinstrumentationtunnelwhichinturn spillsovertothecontainmentfloor.Theinstrumentationtunnelisassumedtobefullandtherefueling cavityisconsideredtobe99%full(48,823gallons)indeterminingtheminimumsumpwaterlevel.

WaterspillsfromtherefuelingcavityintothereactorcavitythroughopenSealRingHatches.Spillover throughtheeight(8)SealRingHatches(eachabout24inchdiameter)directlyentersthereactorcavity andspillsintotheinstrumenttunnelpriortoreachingthecontainmentfloor.SealRingHatchProtective CoversareinstalledovertheopenSealRingHatches(raised8.5inchabovetheopening).Thesecovers allowunimpededairandwaterflowduringplantoperation.TheopenSealRingHatcheswithprotective coversinstalleddonotpresentcrediblelocationsfordebrisblockageduetothelargesizeofthe openings.Nootherminimumflowrestrictionsbetweentherefuelingandreactorcavitiesexist.

5BMPS3,ChemicalEffectsQuestions AECLperformeddissolutiontestsbothwithandwithouttrisodiumphosphate(TSP)inthebeakers.The testingshowedthattheteststhatincludedTSPshowedaninhibitionofthecalciumdissolution.However, fortheheadlosstestingthelicenseestatedthattheyappliedthecalciumquantitydeterminedbythe uninhibited(nonTSP)benchtesting.DatafromthelowestallowablepH(7.0)wasusedwhen determiningtheamountofcalciumtobeaddedtotheheadlosstest.Thecalciumconcentrationusedfor headlosstestingwas14.7mg/L.Thisvalueissignificantlylowerthanthemeasuredvalueforthe30day benchscaledissolutiontesting,whichusedscaledamountsofconcretetorepresenttheMPS3condition.

Pleaseprovidethefollowingadditionalinformationinordertodeterminethatthetestingwasperformed inanacceptablemanner:

14. ThesolubilitydataforcalciumshowsincreaseddissolutionatlowerpHranges.Intable02,,totheDecember18,2008letter,thecalciumconcentrationsforpH5.0and6.0 arelowerthantheconcentrationforpH7.0.Inaddition,page11of30statesthattheconcrete samplesinthebeakertestsfullydissolvedinthepH5.0and6.0testsbutwerenotfullydissolved inthepH7.0and8.0tests.PleaseexplainwhythebenchtestsatlowerpHranges,inwhichthe

Page41of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

concretefullydissolved,resultedinlowerconcentrationsofdissolvedcalciumthanthebench testsathigherpHranges,inwhichtheconcretedidnotfullydissolve.

15. ForMPS3,thecalciumdissolutiontestatpHof7.0resultedina30daycalciumconcentrationof 78mg/L.TheDecember18,2008,letterstatesthatthepH7.0case(withoutTSPpresent)was usedtodeterminetheconcentrationofcalciumintheRigB9test.However,thecalcium concentrationusedforRig89testingwas14.7mg/L.Pleasejustifywhy14.7mg/Lisa representativevalueintheRig89testingwhenthedissolutiontestingconductedwithscaled quantitiesofconcreteresultedinacalciumconcentrationof78mg/L.

16. DNC'stestingwasperformedat104F,whichiswellbelowearlypostIossofcoolantaccident pooltemperatures.Thesolubilityofcalciumphosphate(hydroxyapatite)decreasesasthe temperatureincreases.Pleasediscusswhethermorecalciumphosphateprecipitatewouldhave formedintheRig89testsifthistestwouldhavebeenperformedathighertemperature.Ifmore calciumphosphateprecipitatewouldbeexpectedatahighertemperature,whentheshortterm NPSHmarginisapplicable,pleasejustifywhytheoverallRig89testresultsprovideforan adequateevaluationofchemicaleffects.

17. PleasecomparethetotalamountofaluminumthatispredictedtobereleasedbytheAECL modelwiththatpredictedbytheWCAP16530basemodel(i.e.,norefinementsforsilicateor phosphateinhibition).Discussanysignificantdifferencesbetweentheplantspecificpredictions forthetwomethods,includingtheacceptabilityofthesedifferences.

28BResponsetoMPS3,ChemicalEffectsQuestion14 Thecouponsusedinthesetestsweresmallandsubjecttovariabilityofrockandmortarcontent;thus,it mustbearguedthatthecouponsusedinthepH5and6testscontainedlessmortar(theprimarysource ofcalcium)thanthoseusedinthepH7and8tests.Theslightlylowerconcentrationsattainedinthe pH5and6testsrepresentedthelimitofthecalciumsource(mortar)whileslightlyhigher concentrationswereattainedinthepH7and8tests,despitethecouponsremainingstructurallyintact.

Amoredetailedexplanationoftheapparentconflictbetweentheresultsofthesetestsandcalcium solubilitydataisincludedbelow.

Concreteisinherentlybasic,andexpertsandliteratureagreethatconcretedissolutionratesincreaseas theexposedmediumbecomesmoreacidic.IntheAECLTestReport[D22D]ofbenchtoptestsconducted forDominion,theresultsofdissolutiontestssimulatingtheMPS3concretesurfaceareatovolume ratioF3 FdoshowhigherdissolutionratesatlowerpHranges,buttheultimateconcentrationsreached werelowerinthetestsatpH5and6thaninthoseatpH7and8(XFigure24X).Thisapparent contradictioncanbeexplainedbythesmallsizeofthecouponsused:eachcouponmeasured approximately0.4x1.2x0.5cm,andweresmallincomparisontosimilartestsperformedforMPS2.Asa resultoftheirlimitedsize,twoofthecouponscompletelydissolvedinthepH5and6tests.Aswell, theirsmallsizemadethemmorepronetocontainingnonuniformproportionsofrockandmortar.

Consequently,thecalciumconcentrationsmeasuredtowardtheendofthepH5and6testsrepresent thenaturallimitwhenalloftheconcretehaddissolved.Thedatafromallofthetestswerefitto

3 Asestimatedatthetime;theratiohassincechanged.

Page42of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Equation6.TheearlyplateauseeninXFigure24XforthepH5and6testsbiasedtheirextrapolated concentrationsatt=showninXTable11X(TableO2oftheabovementionedletter).

Equation6 Table11:CalciumConcentrationFittingParametersofEquation6fromMPS3Dissolution Tests Parameter pH5 pH6 pH7 pH8 C[mg/L]

82 77 103 68 k[h1]

0.017 0.0049 0.0029 0.0034

Figure24:CalciumReleaseDatafromMPS3DissolutionTestswithoutTSPat90C

Notethelinesarefitsofthedatasetstoafirstorderreleaseequation.

TherecoveryofcouponsfromthepH7and8tests,andthelackofanyobviousplateauinXFigure24X, stronglyimpliesthattheresultsofthesemoreimportanttestswerenotbiasedbyalimitedcalcium source.TheresultsofthepH7test,inparticular,wereusedinthedesignofreducedscaletest,aspH7 istheminimumallowedsumpwaterpH.However,itshouldbenotedthatthisremainsaconservative estimateofcalciumreleasesincetheMPS3sumpislikelytoremainmainlyabovepH8.

Time (h)

[Ca] (mg/L) 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 0

10 20 30 40 50 60 70 80 90 pH 5 pH 6 pH 7 pH 8

Page43of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

29BResponsetoMPS3,ChemicalEffectsQuestion15 Thevalueof14.7mg/LusedintheRig89testingwascalculatedbyappropriatelyscalingtheresultsof thedissolutionteststomatchupdatedestimatesoftheMPS3concretesurfacearea.Thisresponsewill show:

1. Theconcretesurfaceareatovolumeratiousedinthebenchtopdissolutiontestswasbasedon estimatesoftheconcretesurfaceareathatwerelaterupdated;

2. Theresultsofthedissolutiontestsmaybenormalizedtounitsofcalciumreleaseperunitarea, whichmaythenbeusedtocalculatetheexpectedcalciumreleaseandcalciumconcentrationin MPS3basedontheupdatedconcretesurfacearea;

3. Itisappropriatetousethefittotheentiredatasettodeterminethescaledcalcium concentrationratherthantoscaletheanalysisresultobtainedonday30(78mg/L),whichis moresubjecttosamplingandstatisticalerrors.

Theconcretesurfaceareatovolume(SA/V)ratiousedinthebenchtopdissolutiontestswasroughly6 timesgreaterthanthecurrentcalculatedSA/VratiousingdatafromERC25212ER060013Rev.2[X21X]

andleadstotheapparentdiscrepancy.ThedissolutiontestsconductedfromFebruarytoMarch,2008, usedcouponssizedtomeettheSA/VratiocalculatedfromRev.1ofthatdocument[D23D]andincluded scaledquantitiesoffibrousdebris.XTable12XcomparestheSA/Vratiousedinthedissolutionteststo thosecalculatedfromthesourcereferences.Itisimportanttonotethat,bydesign,thereisno uncoatedconcretewithintheMPS3containmentandthatallvaluesquotedareconservativeestimates ofbareareasexposedeitherbychippingandwearorbyimpactofthebreakjet[X21X].

Table12:ComparisonofDissolutionTestConcreteSA/VRatiotoMPS3Values Source ERC25212ER060013 Rev.1 DissolutionTest ERC25212ER060013 Rev.2 Date 2007/09 2008/02-2008/03 2008/04 Submerged Concrete 1000ft2 (9.29x105cm2) 0.4x1.2x0.5cmcoupons (2.56cm2) 100ft2 (9.29x104cm2)

ExposedConcrete 1932ft2 (1.795x106cm2)

408ft2 (3.79x105cm2)

Volume 3,819,002lbm

@61.55lbm/ft3 (1.757x106L) 4L 160,000ft3 (4.53x106L)

SA/VRatio (Submerged) 0.529cm2/L 0.64cm2/L 0.0205cm2/L SA/VRatio(Total) 1.55cm2/L

0.104cm2/L

Page44of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

BecausetheconcreteSA/Vratioforcontainmentdiffersfromthattested,theresultsobtainedarenon representativebutmaybeappropriatelyscaled.Normalizationofthedissolutiontestdatamaybe performedbydividingtheresults(inmg/L)bytheSA/Vratio(0.64cm2/L),asindicatedbytherighthand verticalaxisinXFigure25X.Similarly,thefittothecalciumconcentrationdata,describedbelow,mayalso benormalizedtoproduceacalciumreleaseequation.Thus,the30daycalciumreleaseperunitareaof concretecanbereadfromthefigureorcalculatedfromthefitandusedtocalculatethecalciumrelease fromaknownsurfaceareaofconcrete.

XFigure25XalsoshowsthecurvefitstothedatarepresentedbyEquations7and8.Thesewere determinedusingrobustfittingprocedureswithinTableCurve2DF Fthatreducethefittingerrorscaused bydataoutliers.TheconstantsfoundwithinEquation7werereportedinTable25ofthebenchtop TestReport[X22X]andTableO2ofDNCsDecember18,2008letter.Equation8maybecalculatedfrom Equation7bydividingtheinitialconstantF4 Fbythetestedsurfaceareatovolumeratio,0.64cm2/L.

Equation7

Equation8

TableCurve2DisproducedanddistributedbySystatSoftwareInc.

4 Theinitialconstant,C,wasdeterminedtobe102.5mg/L,wherethetenthsdecimalplaceshouldnotbe consideredsignificant.

Page45of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure25:CalciumReleaseDatafromMPS3pH7andpH8DissolutionTests(withoutTSP)@

90C

Notethelinesarefitsofthedatasetstoafirstorderreleaseequation.

Itisappropriatetousethefitratherthantherawdatatodeterminethe30daycalciumconcentration, asdriftsinpH,samplingerrors,andstatisticalerrorassociatedwiththeanalysistechnique,ICPOES,may alterthemeasuredconcentration.

After30days,theexpectedcalciumreleaseatpH7and90C:

160 1 exp0.0029* 720140 UsingtheSA/Vratiofromthe4thcolumnofXTable12X,0.104cm2/L,theexpectedcalciumconcentration is:

F5 F

140

• 0.104 14.6 Forcomparison,theexpectedcalciumconcentrationatpH8is10.1mg/Lbysimilaranalysis.

5 Withintheerrorofthisanalysis,thereisnosignificantdifferencebetweenthisresultandthepreviously reportedvalueof14.7mg/L.

Time (h)

Time (d)

[Ca] (mg/L)

Ca Release (mg/cm

2) 0 100 200 300 400 500 600 700 0

2 4

6 8

10 12 14 16 18 20 22 24 26 28 30 0

10 20 30 40 50 60 70 80 90 0

16 32 48 64 80 96 112 128 144 pH 7 pH 8

Page46of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

ThisresultmaybecomparedtotheWCAP16530methodofcalculatingcalciumrelease,asdescribedby Laneetal[X8X].Inutilizingthismethod,thecalculatedpHhasbeenused;inordertomaximizethe releaserate,themaximumpHwasusedtocalculatethereleasefromTranscoThermalWrap(around pH8.1)andtheminimumpHwasusedtocalculatethereleasefromconcrete(aroundpH8.0).Bythis method,thecalculatedcalciumreleasefromconcreteisminisculeF6 F;mostofthecalciumreleasedcomes fromfibrousdebris.Thecalciumconcentrationispredictedtoplateauat10.6mg/L(XFigure26X),the saturationlimitofcalciumreleasedfromTranscoThermalWrapatpH8.1and165.34F(74.08C).

Therefore,thecalciumconcentrationobtainedbyscalingtheAECLpH7dissolutiontestresultsis conservativewithrespecttotheWCAPresult.

Figure26:CalciumReleasefromMPS3FibrousDebris/ConcreteIAWWCAPMethod

30BResponsetoMPS3,ChemicalEffectsQuestion16 Potentiallyincreasedcalciumphosphatesolubilityathighertemperaturesdoesnotsignificantlyimpact theMPS3testresultsduetosignificantconservatismsbuiltintothetestingprogram.

1. ThereisnosignificantsourceofcalciumintheMPS3containment.Theonlypotentialcalcium sourcesforMPS3containmentareuncoatedconcreteanddislodgedfibrousinsulation.By design,thereisnouncoatedconcreteintheMPS3containment.FortheRig89testing,atotal

6 WhentheTranscoThermalWrapcontributiontocalciumreleaseisneglected,thecalculatedcalcium releasefromconcreteusingtheWCAPmethodislessthan5g.Bycontrast,whentheTranscoThermalWrap contributionisincluded,thecalculatedcalciumreleaseisnearly50kg.

Time (h)

Time (d)

[Ca] (mg/L)

Ca Release (kg) 0 100 200 300 400 500 600 700 0

2 4

6 8

10 12 14 16 18 20 22 24 26 28 30 0

2 4

6 8

10 12 0

10 20 30 40 50

Page47of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

of508ft2ofconcreteisassumedtobeuncoatedincontainment.Ofthattotal,308ft2is considereduncoatedduetothebreakjetimpactingcoatedwalls.Theremaining200ft2is marginfordamagedconcretecoatingincontainment.Nocalciumsilicateinsulationexistsin containmentatMPS3.Calciumreleasesduetodegradationofotherdislodgedinsulationare includedinthetotalcalciumreleaseusedinthetesting.Basedontheconservativeestimatesof existinguncoatedconcrete,therewillbesignificantlylesscalciumreleasedintothecontainment sumpwaterthanwastested.

2. Inthebenchtoptesting,TSPinhibitedcalciumreleasefromuncoatedconcrete.Identicaltests wereruninthebenchtoptestingtodeterminetheeffectofTSPoncalciumconcentration.

Bothsetsoftestswereconductedwithscaledamountsofconcreteandfibrousinsulation.In onesetoftests,noTSPwasused.Inanidenticalsetoftests,arepresentativeconcentrationof TSPwasestablishedinthetestwater.AtpH7,theexpectedcalciumconcentrationin containmentintheabsenceofTSPis14.6mg/L.InthepresenceofTSP,the30daycalcium concentrationduringthebenchtoptestingwas2.2mg/L.IntheabsenceofTSP,theconcrete couponsinthetestshowedsignificantdissolution.WhenthetestswererepeatedwithTSP present,concretecouponsinthetestshowednoevidenceofdissolutionandexperiencedless thana1%lossinmass.Forconservatism,theresultsfromcalciumdissolutiontestswithoutTSP presentwereusedtodeterminetheamountofcalciumtoaddtotheRig89testtank.

3. Concreteusedintestingwasnotsafetyrelatedconcreteandthuswasmorelikelytodegradein thebenchtoptestingthanisthesafetyrelatedconcreteinstalledincontainment.

4. ConcretedissolutiondataforpH7wasusedinthetestingtodeterminetheamountofcalcium releasedandtheamountofcalciumusedinchemicaleffectstesting.ThepHintheMPS3 containmentwaterisexpectedtobeabove8.0followingtheLOCAresultinginmuchless calciumrelease.ConcretedissolutionislowerathigherpH.Expectedlongtermcalcium concentrationatpH8(withoutTSP)is10.1mg/Lascomparedtotheexpected(andtested) calciumconcentrationatpH7(withoutTSP)of14.6mg/L.Thus,thecalciumconcentrationin containmentislikelytobeasmuchas30%lowerthanthetestedvaluedueonlytothepHin containment.

5. Atotalof14calciumadditionsweremadetotheMPS3Rig89test.Theseadditionshada minimalimpactonheadlossthoughTSPwaspresentinthetesttankattheexpected concentrationincontainment.ThisTSPconcentrationfarexceededtheamountneededto precipitatealloftheavailablecalciuminthetest.Thefirstcalciumadditionwasmadetogether withanaluminumadditionandtheheadlossincreasedfrom0.43to0.73psig.Theremaining 13calciumadditions(allmadeseparatelyfromaluminumadditions)hadnosignificantimpact onheadloss.

Page48of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Theaboveinformationdemonstratesthatsufficientconservatismexistsinthedeterminationofpost LOCAsumpwatercalciumconcentrationtooffsetthepotentiallowersolubilityofcalciumatthehigher postLOCAsumptemperaturesexpectedearlyintheaccident.

31BResponsetoMPS3,ChemicalEffectsQuestion17 TheWCAP16530basemodelisanempiricalmodelofthealuminumreleaserate(RR)basedonthedata setdescribedbyLaneetal[X8X],whichincludeddatafromICET1,CR6873,WCAP7153Aand WCAP16530.TheWCAPmodelisdescribedbytheEquation9andtheresultsareshowninXFigure27X.

Equation9 Figure27:3DIllustrationoftheWCAPAluminumReleaseModel

TheAECLmodelisasemiempiricalmodelofthealuminumreleaserate,inthattheequationformwas developedfromfirstprinciplesbuttheparameterswerefittoliteraturedata.Thereleaseequation takesanArrheniusformwithtemperatureand,sincethecorrosionreactioninvolveshydroxide,the releaserateislikewiserelatedtotheexponentialofthepH.Thedatasetusedtofitthemodelwas describedbyGuzonasandQiu[D24D]andwasverysimilartothatusedfortheWCAP16530model.The AECLmodelisdescribedbyEquation10andtheresultsareshowninXFigure28X.

Equation10

Page49of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure28:3DIllustrationoftheAECLAluminumReleaseModel

BothmodelsignoreanytimedependenceoftheAlreleaserate.Asonemightexpect,thetwomodels givesimilarpredictions.Mathematicalcomparisonofthetwomodelsshowsthattheydiffermainlyat temperaturesabovethenormalboilingpointofwater.TheWCAPmodelpredictshigherreleaseat moderatepHvalues(betweenpH79.5)andlowerreleaseathighpHvalues,asshowninXFigure28X.At moremoderatetemperatures,thetwomodelspredictverysimilarreleaserates.Forexample,ICET Test5[X9X]wasconductedat60CatpH8.08.5,andbothmodelsareobservedtoconservativelypredict thelongtermaluminumrelease,especiallywhenreleasefromsprayedaluminumwithhighpHsprayis included(XFigure29X).

Page50of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure29:3DDifferentialofWCAPandAECLAluminumReleaseModels

Figure30:WCAP/AECLAluminumReleaseModelofICETTest5AluminumConcentration

NoteICETTest5concentrationdataadaptedfrom[X9X].SpraypH,reportedas<12,wastakentobe11 forcalculations.

TheMPS3postLOCAsumpandsprayoperatesmainlyintherangeofpH8.08.5,wheretheWCAP modelpredictsagreateraluminumreleaserateathightemperaturesthantheAECLmodel(XFigure31X).

Time (d)

[Al] (mg/L) 0 3

6 9

12 15 18 21 24 27 30 0

50 100 150 200 250 300 350 Submerged Al Only Submerged and Sprayed Al AECL Model WCAP 16530 Model ICET Test 5

Page51of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Forthe1080ft2ofsprayedand120ft2ofsubmergedaluminumreportedtobepresentatMPS3[X21X],

theWCAPmodelpredicts14.6kgAlwhereastheAECLmodelpredicts8.15kgAl(XFigure31X).Notethat thescaledequivalentof7.6kgAlwasaddedduringtheRig89testF7 Fandthatthelast2aluminum additions(i.e.,additions11and12,XFigure32X),representingover30%ofthealuminumadded,didnot produceincreasesinheadloss,suggestingaheadlossplateau.Althoughslightlymorealuminumwas neededtomeetthepredictedaluminumrelease,theobservedheadlossplateauallowsconfident predictionoftheheadlossforthepredictedaluminumrelease.

Figure31:ComparisonofAECL/WCAPAluminumReleaseModeloftheSubmerged,Sprayed andTotal(Combined)AluminumReleaseforMPS3PostLOCAContainment

7Althoughthescaledequivalentof7.6kgofAlwasaddedduringthetest,only7.45kgcanbesaidtohave precipitatedwithcertaintyduetotheerroruncertaintyresultingfromthemethoddetectionlimitforICPOESfor aluminum(0.4mg/L)

Time (d)

Al Release (kg) 0 2

4 6

8 10 12 14 16 0

5 10 15 20 25 30 0

5 10 15 20 25 30 AECL WCAP Submerged Sprayed Total

Page52of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

Figure32:Rig89HeadLossTraceCorrectedtoMatchApproachVelocityofMPS3

Without30dayaluminumcorrosiontestswheretemperatures(andpressures)oftheMPS3sumpare simulated,itisdifficulttospeculateonthesignificanceofthedifferencebetweenpredictionsofthe WCAPandAECLmodels.TheonlyavailabledataforaluminumreleaseatpH8fortemperatures exceedingthenormalboilingpointofwaterwasreportedfora90minutetestat265F(129C)byLane etal[X8X];thereportedreleaserateof6.6mg/(m2s)wasmanytimesgreaterthanthatpredictedby eithermodel(theWCAPmodelpredicts2.7mg/(m2s),andtheAECLmodelpredicts1.0mg/(m2s)).

Whilethiscomparisonmayseemtohighlightapparentdeficienciesinbothmodels,thedeficienciesof thedatasetaremoreapparent,asitcannotbesaidwithanycertaintythatthevalueof6.6mg/(m2s)is eitheraccurateorrepeatable.Therearemanyvariablestocontrolincorrosiontests,anditisdifficultto getconsistentresults;hence,Laneetal[X8X]couldmeasureareleaserateof0.75mg/(m2s)atpH8and 190F(88C)whileotherscouldmeasurelowerratesatmoresevereconditions:Reidetal[X10X]

measured0.13mg/(m2s)atpH8and200F(93C),Belletal[X11X]measured0.20mg/(m2s)atpH8and 210F(99C),andJainetal[X12X]measured0.53mg/(m2s)atpH10and194F(90C).Thesevaluesare comparedtoWCAPandAECLmodelpredictionsatpH8inXFigure33X.Itisclearthereisalargescatterin thetestdata,withtwodatapointsclusteredcloselytogetherandoneverymuchhigher.Thismay reflectdifferencesintestmethodologyorconditions;AECLhasfoundexperimentaluncertaintiesof about30%innominallyidenticaltests.Bothmodelspredictreleaserateswithinthescatterofthe plotteddata;theAECLmodelbetterfitsmostofthedata,buttheWCAPmodelmorecloselymodelsthe averagevalueandisthemoreconservative.However,thelimitedexperimentaldataavailabledonot 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 29-May 05-Jun 12-Jun 19-Jun 26-Jun 03-Jul 10-Jul 17-Jul 24-Jul Corrected Head Loss (psi) 1 2

1 2

Stoppage Events Power loss.

Power loss.

1st Al Addition 1st Ca Addition 2nd Ca Addition 3

4 5

6 7

8 9

2nd Al Addition Ca Additions 3

4 10th Ca Addition 5

6 7 8 9 10 11 12 11 12 13 14 Ca Additions Al Additions

Page53of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

provideabasisforselectingonemodelovertheother,andnosignificancecanbeascribedtothe differencesinthepredictedaluminumrelease.

Figure33:ComparisonofAECLandWCAPAluminumReleaseModelPredictionsand MeasuredValuesatpH8

Itshouldalsobenotedthatneithermodelwasdevelopedtopredictshorttermreleaserates.Although shorttermreleaseratesmaybehigherthanpredictedbythemodels,longtermreleaseratesarelikely tobelowerthanpredicted,asindicatedbytheresultsofICETTest5(XFigure30X)andothertestsshowing aplateauinreleaserates,includingtheclassicaluminumcorrosiontestsdescribedbyTroutner[X13X,X14X].

6BReferences

1FisherNJ,BartlettMM,ChengQ.ReducedscaletestingforMillstone2replacementcontainmentsumpstrainers.

AECLTestReportMIL234325TR001,Rev.0;2006September.

2FisherNJ,BartlettMM,ChengQ.LargescaletestingforMillstone2replacementcontainmentsumpstrainers.

AECLTestReportMIL234325TR002,Rev.0;2006October.

3LeggM.MillstoneUnit2inputsforGSI191strainerheadlosstesting.DominionERCNo.:25203ER070029,Rev.

0;2007December20.

4GuzonasD,EdwardsMK,ChengQ,HaqueZ,DeadmanJ.Finalreportonstrainerdebrisbedheadlossarisingfrom prototypicalchemicaladdition:Dominion:Millstone2.AECLReportNo.:MIL234325TR004Rev.1;2009October.

Temperature (ºF)

Temperature (ºC)

Al Release Rate (mg/(m2*s))

180 185 190 195 200 205 210 215 220 120 122.5 125 127.5 130 132.5 135 137.5 140 0

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Lane et al.

Reid et al.

Bell et al.

WCAP AECL

Page54of54 DraftofInformation(DOI)-ThisDOIisentirelypreliminaryandusedsolelytosupportteleconferencingtoaffirma commonunderstandingofthescopeandintentofseveralNRCquestions.ThisDOIprovidesdraftresponsestothe requestsforadditionalinformationand,accordingly,theinformationprovidedshouldnotbeusedaspartofany NRCfinaldecisionmakingregardingthistopic.

5LeggM.MillstoneUnit2inputsforGSI191chemicaleffectstesting.DominionERCNo.:25203ER060007,Rev.

3;2008Apr15.

6LeggM.MillstoneUnit2inputsforGSI191chemicaleffectstesting.DominionERCNo.:25203ER060007,Rev.

1;2007Aug21.

7GuzonasD,QiuL,EdwardsM.Resultsofbenchtopchemicaleffectstests:Dominion-Surry1and2,NorthAnna 1and2,Millstone2and3.AECLReportNo.:DOM34325TR001Rev.0;2008July.

8LaneAE,ByersWA,JackoRJ,LahodaEJ,ReidRD.Evaluationofpostaccidentchemicaleffectsincontainment sumpfluidstosupportGSI191.WestinghouseReportNo.:WCAP16530NPA;2008March.

9DallmanJ,LetellierB,GarciaJ,MadridJ,RoeschW,ChenD,HoweK,ArchuletaL,SciaccaF,JainBP.Integrated chemicaleffectstestproject:Consolidateddatareport.NUREG/CR6914;2006August.

10ReidRD,CrytzerKR,LaneAE.EvaluationofadditionalinputstotheWCAP16530NPchemicalmodel.

WestinghouseReportNo.:WCAP16785NP;2007May.

11BellMJ,BulkowskiJE,PiconeLF.Investigationofchemicaladditivesforreactorcontainmentsprays.

WestinghouseReportNo.:WCAP7153A;1975April.

12JainV,HeX,PanYM.Corrosionratemeasurementsandchemicalspeciationofcorrosionproductsusing thermodynamicmodelingofdebriscomponentstosupportGSI191.U.S.NRCReportNUREG/CR6873;2005April.

13TroutnerVH.Observationsonthemechanismsandkineticsofaqueousaluminumcorrosion:Part1roleofthe corrosionproductfilmintheuniformaqueouscorrosionofaluminum.Corrosion.1959;15(1):9t15t.

14TroutnerVH.Uniformaqueouscorrosionofaluminumeffectsofvariousions.HanfordAtomicProducts OperationReportNo.:HW50133;1957June10.

15Fisher,N.J.,Cheng,Q.andHaque,Z.,ReducedScaleTestingforSurry1and2ReplacementContainmentSump Strainers,AECLTestReportSUR234325TR001,Rev.0,2007November.

16Fisher,N.J.,Bartlett,M.M.andCheng,Q.,ReducedScaleTestingforMillstone3ReplacementContainment SumpStrainers,AECLTestReportMIL334325TR001,Rev.0,2007March.

17Manson,R.E.,Ophel,I.L.,ExperiencewithSlimeFormingOrganismsinChalkRiverReactorSystems,AECL Report1124,1960September.

18Fisher,N.J.,Cheng,Q.andHaque,Z.,ReducedScaleTestingforNorthAnna1and2ReplacementContainment SumpStrainers,AECLTestReportNAN234325TR001,Rev.1,2008February.

19Cheng,Q.,Fisher,N.J.,Bartlett,M.M.,Haque,Z.,LargeScaleTestingforMillstone3ReplacementContainment SumpStrainers,AECLTestReportMIL334325TR002,Rev.0,2007April.

20EdwardsMK,ChengQ,GuzonasD,HaqueZ,Finalreportonstrainerdebrisbedheadlossarisingfrom prototypicalchemicaladdition:Dominion:Millstone3.AECLReportNo.:MIL334325TR004Rev.1;2009October.

21LeggM.MillstoneUnit3inputsforGSI191chemicaleffectstesting.DominionERCNo.:25212ER060013,Rev.

2;2008Apr16.

22GuzonasD,QiuL,EdwardsM.Resultsofbenchtopchemicaleffectstests:Dominion-Surry1and2,North Anna1and2,Millstone2and3.AECLReportNo.:DOM34325TR001Rev.0;2008July.

23LeggM.MillstoneUnit3inputsforGSI191chemicaleffectstesting.DominionERCNo.:25212ER060013,Rev.

1;2007Sep24.

24GuzonasD,QiuL.ChemicaleffectstestingforMillstoneUnit2.AECLReportNo.:MIL234325401000;2008 March.