Uranium-233 Hanford Reach

Riverbed

By: Norm Buske The RadioActivist Campaign search@igc.org

Rev. 2

January 2006

in the

�

Uranium-233 in the Hanford Reach Riverbed

Summary The 50-mile Hanford Reach of the Columbia River in Eastern Washington is recognized as

• a scenic recreation area and national monument.• crucial spawning habitat for wild chinook salmon.• part of the Columbia Basin irrigation project and the mid-Columbia hydro-

electric power generation system.• collecting radioactive and toxic wastes from old nuclear weapons production

at the Hanford Site.

ThepresentstudyconfirmsthatsomeofHanford’sworstradioactivewasteisstickingtoColumbiariverbed sediments.

Hanfordreactorsproduceduranium-233(U-233)fortacticalnuclearweaponsduringtheColdWar.TRAChassurveyedHanford’sradioactivityintheriverbedsince1999.BasedontheevidenceinTRAC’sstudies,theWashingtonStateEcologyDepartmentjointlysampledriverbedsedimentswithTRACon23 September 2004. Ecology then analyzed colloids from the sediment samples to test for U-233. The objectivewastoanswerthequestion:IsU-233ofHanfordoriginpresentorabsentinHanfordReachsediment?

TRACandEcologyagreedonsamplelocations.OnebackgroundsamplewascollectedatVernita,upstreamofHanfordandsonotinfluencedbyHanforddischargesintotheriver.Threesampleswerecollectedfromtheriverbed,neartheupstreamendofD-Island.ThosethreeD-Islandsamplesprovidedthetest data for this study. Unfortunately,thedetectionleveltoidentifyU-233wasnotclearlyspecifiedforEcology’sanalyses.Ecology’sICP-MSanalyseshadadetectionlevelahundredtimesabovethelevelrequiredfordetectionofU-233andsoprovidedonlynilresults.Ecology’salphaspectrometrydidnotreportU-233distinctfromnaturally occurring U-234.

EcologyprovideditsalphaspectrometrydatafilestoTRAC.TRACappliedanordinaryanalyticalprocedure to separate a U-233 peak from the U–234 peaks in the spectra. The three test samples from D-IslandyieldedU-233intherangeof0.010 to 0.024 pCi/g.Thebackground(VernitaBKG),blank,andspikeresultswereeach0.000pCi/g.These results confirm the presence of U-233 of Hanford origin, on sediments in the riverbed of the Hanford Reach. Thelevelofconfidenceinthisconfirmationdependsonpreviously held beliefs.

TheU-233wasteispresentincolloids that stick on sediments under the Hanford Reach. The colloidsinteractchemicallywithwaterintheriverbed,wheresalmonhatchlingslive.

Contents Summary...................................................................................................BelowBackground and Sampling Locations ........................................................ 2Introduction................................................................................................ 3Procedure:Chronology............................................................................. 4Analytical Results ...................................................................................... 7Confidence................................................................................................ 8Implications............................................................................................... 9References(hyperlinked)..........................................................................10ProducedandPublishedby......................................................................10

�

Uranium-233 in the Hanford Reach Riverbed

Reactor areas are identified by letter. Major salmon spawning areas are identified by numbered black bars (Areas #1 - #10). The Vernita background sample was collected a half mile upstream of Hanford River Mile (HRM) zero at Vernita Bridge. The three test samples were collected from the upstream end of D-Island, about HRM 10.5.

�

Uranium-233 in the Hanford Reach Riverbed

Introduction Wildfallchinooksalmonspawnintenareasthatunderliethe50-mileHanfordReachoftheColumbiaRiverinEasternWashington;seeLocationsonPage2.Theseareasarecrucialhabitatforthewildsalmonhatchlings(alevin) that are believed to maintain the strength of the salmon stock for a billion-dollar-a-year commerce.

BeginninginWorldWarII,Hanford’snuclearreactors,builtontheshoreoftheColumbia,usedriverwatertocooltheircores.Radioactiveandtoxicwastesfromthereactorsweredischargedbackintotheriver.

Hanford’sbestknownnuclearproductwasplutonium-239(Pu-239).Pu-239poweredtheTrinityTestatAlamogordo,NewMexico,on16July1945andthebombdroppedonNagasaki,Japanon9August1945,endingWorldWarII.PlutoniumfromHanford’sreactors provided much of the strength of the U.S. nuclear deterrent during the Cold War.

InadditiontoPu-239,Hanfordproducedawidevarietyofspecial nuclear materials, ranging from test specimens to full production.DuringtheColdWar,allHanford’s9reactorsproduceduranium-233(U-233)fortacticalnuclearweapons.

In2000,theU.S.DepartmentofEnergy,owneroftheHanfordSite, began to declassify information about its still semi-secret productionofU-233inHanford’sreactors.Followingarequest

undertheFreedomofInformationAct,TRAClookedovertensofthousandsofpagesofdocumentsrelatingtoU-233production,processing,andwastedisposalatHanford.SeeReferences4and5,forasummaryofTRAC’sdiscoveries.

Since1999,TRAChassurveyedartificialradioactivityinthesedimentsundertheHanfordReachoftheColumbiaRiver.SeeReferences3,4,and5.BasedonTRAC’saccumulatingevidenceofU-233wasteintheriverbed,theWashingtonStateEcologyDepartmentjointlysampledriverbedsedimentswithTRACon23September2004.Thesamplesweresplitthreewaysforqualitycontrolandassurance.

TRACextractedthecolloidalandfinefractionfromonesetofthesamplesplitsandreturnedthatfraction to Ecology. Ecology then analyzed those four sample fractions, as Stakeholder Columbia River Samples,totestforthepresenceorabsenceofU-233.Theobjectivewastoanswerthequestion:Is U-233 of Hanford origin present in Hanford Reach sediment?—Thiswasaconfirmationorrefutationtest,notan indication of the importance of any U-233 that might be present.

RepresentativesofEcology,ofthestateDepartmentofHealth(DivisionofRadiationProtection),ofthePacificNorthwestNationalLaboratory,andofTRACparticipatedinsamplecollectionon23September2004.Theseparticipantscollaboratedtoselecttheexactsamplinglocations.Thisassuredthatsamplelocationwasunobjectionablefromanyknownstandpoint.OnesamplewascollectednearVernita,upstreamofHanfordoperationsandsowasnotinfluencedbyHanforddischargesintotheriver.Thatsamplerepresentedbackground(BKG)forthisstudy.Threesampleswerecollectedfromtheriverbed,neartheupstreamendofD-Island.ThosethreeD-Islandsamplesprovidedthetestdataforthisstudy.

Therequireddetectionlevelwasdiscussed,butevidentlynotagreed.Ecologyunderstoodthattheircontractlaboratorieswould“runthesamplesatthelowestdetectionlimitforwhichtheyarequalityassured/qualitycontrolled[Reference1].”TRACunderstoodthatdetectionleveltobe1part/trillion(=0.1µg/kg)bymassor0.01pCi/gactivity,thelevelrequiredtodetectU-233.Ecology’sactualdetectionlevelwas100parts/trillion(=10µg/kg),byICP-MS.Ecology’sICP-MSanalyseshadadetectionlevelahundredtimesabovetherequiredlevelandsocouldnotprovideanymeaningfulresults.

D-Reactor is one of Hanford’s 9 nuclear reac-tors that produced U-233 for nuclear weap-ons during the Cold War.

�

Uranium-233 in the Hanford Reach Riverbed

Ecology’salphaspectrometrydidnotreportU-233separatelyfromU-234intheirisotopicuraniumanalyses.

The alpha spectra of the samples and reference materials, for isotopic uranium, contain alpha energypeaksfornaturallyoccurringU-234,U-235,andU-238,aswellaspeaksforartificialU-233ifitispresentintheanalyzedmaterial.TheU-233alphapeaksarewithinthegeneralregionoftheU-234peaks.Therefore,Ecology’scontractlaboratorydidnotreportU-233andU-234separately,butinsteadreportedthesumofthosetwoisotopicactivitiesas“U-233/U-234”.

Nonetheless, the main alpha emission from U-233 has distinctively higher energy than the highest alpha energy from U-234 decay. Therefore, the activity of U-233 can be analytically separated from the activityofU-234,eventhoughEcology’scontractlaboratorydidnotmakethisdistinctionintheirreport.

TRAC has performed this ordinary, post-analysis separation of U-233 from U-234. This is the report of that analytical separation of the three test samples, the background sample, a tracer, and a blank.

Procedure: Chronology Onebackground(BKG)samplewascollectedahalfmileupstreamofVernita,whichisupstreamofHanfordSiteoperations.Thislocation(onWGS84datum)isausualreferencebackgroundforHanfordSite impact studies:

VernitaBKG: NorthLatitude: 46.63435° WestLongitude: 119.75458°

TheVernitasampleprovidedacheckforanon-HanfordsourceofU-233,suchasworldwidefalloutfromatmospherictestingofnuclearweaponsorafalsepositiveanalyticalresult.

Threesedimentsampleswerecollectedfromtheriverbed,ontheupstreamendofD-Island.

ThelocationsoftheseD-Islandsampleswereasfollows:

D-Island“1”: NorthLatitude: 46.70151° WestLongitude: 119.54201°D-Island“2”: “ 46.70153° “ 119.54167°D-Island“3”: “ 46.70175° “ 119.54136°

ThosethreesampleswerecollectedupstreamoftheD-/DR-Reactordischargepipestoeliminatethepossibility of sampling U-233 debris that might have been discharged into the river from failed target rods in D-orDR-Reactor.TheD-Reactordischargepipewasmarkedbyriserventslocatedbetween:

northriser: NorthLatitude: 46.70245° WestLongitude: 119.54135°southriser: “ 46.70224° “ 119.54117°

TheDR-Reactordischargepipewasfartherdownstream.

Ateachsamplinglocation,largegravelsandcobbleswereremovedbyhandandshovelfromthesediments.

Finersedimentswereshoveledintoa2-mmstandardstainlesssteelsieve,overacollectingpan(seephotosidebaronnextpage).Thecontentsofthecollectingpanwerepouredintoalargesamplebag.Theparticipantsrepeatedtheprocessuntilthesamplebagwasfilledforthree-waysplitting:toEcology,tothestateHealthDepartment’sDivisionofRadiationProtection,andtoTRAC.TRACpreparedandre-submitted the samples to Ecology.

�

Uranium-233 in the Hanford Reach Riverbed

TRACextractedcolloidsfromthesurfacesofthesedimentsamples,asfollows:About750gofsedimentwereplacedintoroundedsquareplasticbottles(Cole-ParmerCat.No.A-62270-00)whichwerethenfilledwithdistilledwater.Thebottleswererotatedhorizontallyat6rpmforatleast5hours(Cole-ParmerRoto-TorqueRotator,Cat.No.A07637-00withatesttubebasketmodifiedtorotatetwo,quartbottlesatatime).

The rotated samples werethenflushedwithdistilledwaterthroughastandard0.475 mm stainless steel sieve. Theextractedcolloidsandtheveryfinesedimentsthatpassedthroughthesievewereovendriedat<90°Candweighed.On 7 October 2004, TRAC shipped each sample (one of 3 ounces and three of 4 ounces) to Ecology for analysis, along withaChainofCustodyletter.

On28October2004,Ecologyshippedsplitsofthefourpreparedsamples,alongwithaChainofCustodyandanAnalyticalRequesttoParagonAnalytics.Ecology’sanalyticalrequestincludeduraniumisotopes(“ISOUR”)witharemarkforuranium-233(“UR233”).

ParagonAnalyticsusedtheirstandardProcedurePASOP14R9forisotopicuraniumanalysis:“Inthisreport,anyuraniuminthisregionofinterest[inthealphaenergyspectrum]willbereportedasU-233/234[ParagonLaboratoriesCaseNarrative(7December2004)].”Thesampleanalyseswerecompletedon2December.Theresultsincludedaduplicate(“Dup.”)oftheD-Island“1”sampleanalysis,aMethodBlankanalysis,andaMethodTraceranalysis.

AlthoughProcedurePASOP14R9doesnotseparateU-233fromU-234,Paragon’salphaanalysisissharpenoughtoallowcompleteseparationofthe4825keValphaemissionofU-233decayfromthe4776keVemissionofU-234decay.InordertoseparateU-233fromU-234inEcology’salphaspectra,TRACrequestedthespectralfilesfromEcologyon30January2005.Ecologyprovidedthosedatafilesby23March.

Each alpha spectrum consists of a list of counts in each of 508energychannels.Eachspectrumisdocumentedwiththesampledescription;spectrumacquisitioninformation;andcalibrationinformationincluding(bychannel)energy,efficiency,andsharpness(FWHM).Thisinformationallowsareviewertoreplicateanyofthecomputer-generatedanalyticalreportsofradioactivity. Uranium-233hasadistinctalphaemissionpeakat4825keV,withanintensityof84.4%[D.R.Lide,ed.,“CRCHandbookofChemistryandPhysics,”84thed.,CRCPress,BocaRaton(2003)11-184].Thenearestnaturaluraniumpeakisat4776keV,arisingfromU-234decay.TRACselecteda50keVwide,spectralregionofinterest,justabove4800keV.TRACpickedtheparticularchannelsbasedontheenergycalibration information for each alpha spectrum. Then TRAC read the number of counts in each of those

Joint Sampling Process

Step 2: Shovel finer sediments into a 2-mm stainless steel sieve (on right).

Step 3: Empty sieved sediments from collec-tion pan into sample bag. The sample is then divid-ed among the sampling participants.

Rotating two bottles with sediment samples and distilled water to extract colloids from the sedi-ment surfaces.

Step 1: Select a sample site and remove large cobbles and grav-el. This is one of the three sites where the samples were collect-ed upstream of D-Island.

�

Uranium-233 in the Hanford Reach Riverbed

channels.TRACusedthesameprocedureasthelaboratoryusedforU-233/U-234,U-235,andU-238analysis,withtheintensityforthe4825keVpeakofU-233toseparateU-233fromU-234.

TRACprovidedtoEcologythefinalbreak-outofU-233on25March2005,fortheblank,tracer,andtheHanfordsamples,includingthebackgroundsamplefromVernitaandaduplicateanalysis,providingqualityassurance.SeetheAnalyticalResultsonthenextpage.

�

Uranium-233 in the Hanford Reach Riverbed

A

naly

tical

Res

ults

— U

-233

sep

arat

ed a

naly

tical

ly fr

om “

U-2

33/2

34”

in a

lpha

spe

ctra

inParagonAnalytics“IsotopicUranium

ByAlphaSpectroscopyRaw

DataReport”(7Decem

ber2004).

Sym

bol:

N

m

E

Y

A 2s

S

ampl

e

Sta

rt E

nerg

y

AliquotM

ass

Bas

e C

hem

ical

Act

ivity

Unc

erta

inty

Nam

eFile

[keV

]ChannelsCounts

[g]

Efficiency

Yield

[pCi/g]

[pCi/g]

MethodTracer

URB225L4804.181-185

02.

0.3163

0.754

0.00

0 --

MethodBlank

UB225B

4803.179-183

02.

0.3163

0.754

0.00

0 --

Ve

rnita(B

KG)

UA2541

4802.180-184

02.02

0.2839

0.847

0.00

0 --

D-Is

land

“1”

UA2542

4802.180-184

32.

0.3141

0.815

0.01

0±0.012

D-Is

land

“1”Dup.UA2542D4802.179-183

72.01

0.3015

0.845

0.02

4±0.018

D-Is

land

“2”

UA2543

4806.181-185

52.01

0.2976

0.766

0.01

9±0.017

D-Is

land

“3”

UA2544

4807.180-184

32.01

0.3107

0.804

0.01

1±0.012

Foreachsample:

“StartEnergy”isthelowestenergyinthefirstchannelinthesamplespectrumFile,entirelyabove4800keV.S

pectral

channelsabovethisenergyareinsensitivetothehighestenergyem

issionofU

-234at4776keV.Thisdistinguishes

the4825keV

alphaemissionofU

-233decayfrom

theclosestemissionfrom

U-234decay.

“Channels”isthecountingregionthatcom

prisesthefirstfivechannelsbeginningabove4800keV

ineachsample

spectrumFile.E

achchannelisabout10keVwide.The4825keValphaemissionofU

-233isthuscenteredcloseto

the

mid

dle

of th

e 5

chan

nels

of r

egio

n co

unte

d fo

r U-2

33.

N=N

umberofcountsin“C

hannels”countingregioninthesamplespectrumFile.

Activity(A

)ofeachsampleiscalculatedasfollows:

A=(27pCi-seconds/counts)xN/T/m

/I/E/Y

where:T=LiveCountTime=18,000seconds

and:

I=Intensityof4825keValphaemissionofU

-233=0.844

Uncertainty(2s)oftheactivity(A

)ofeachsampleiscalculatedasfollows(nom

inally,95%

confidenceinterval):

2s=TwoStandardDeviationsofN

countswithinthesamplecountingregion

(withnilblankandbackground)=2A/N

1/2

Toconvertµg/kgofU-233topCi/g:multiplyby9.63

�

Uranium-233 in the Hanford Reach Riverbed

Confidence The isotopic separation procedure for the Analytical Results of this report is an ordinary analytical practice.However,thisanalysisisnotpartoftheanalyticallaboratory’sstandardProcedurePASOP14R9for isotopic uranium analysis.

ThereissomeconcernthatthefewalphacountsintheU-233regionofspectralinterestmightbenon-random, false positives. Besides U-233, the other radionuclide having intense alpha emissions in the 4800to4850keVenergyregion-of-interestreportedintheAnalyticalResultsis

Thorium-229: 4814keV(9%intensity),and4845keV(56%intensity)

Thorium-229(Th-229)isthefirstradionuclideproductofU-233decaychain.Ecology’scontractlaboratoryforalphaspectrometryusesTh-229asaquantitativespikeinitsisotopicthoriumanalyses.Therefore,Th-229isaplausiblelaboratorycontaminantinsamplepreparationorotherprocessesinthelaboratory.TherearescenariosinwhichthepositivesamplecountsintheAnalyticalResultsmightbeattributedtoTh-229insteadofU-233.Thesescenariosareprecludedbythezerocountsintheregion-of-interest in the background sample, tracer, and blank spectra.

ThereliabilityoftheconfirmationofU-233inHanfordriverbedcolloidsreportedheredependsprimarily on zero (N) counts in each of

tracer blank background(VernitaBKGsample)

Thesezerosweighheavilyagainstthecountsforthethreesamplesandtheduplicatebeingfalse.Thesethreezerosprovidecumulativeconfidenceofthevalidityofthepositivesamplecountsinthethreetestsamples, ranging from 3 to 7 counts.

The Ecology Department does not presently accept any conclusion regarding the presence or absenceofU-233initsalpharesults,becauseitscontractlaboratory’suraniumisotopicanalysisisnotcertifiedforU-233.Ecologyhasaskeditscontractalphalaboratorytoundertakeaspecialstudyin2006to“showthe‘peaks’forU-233inaspikedsampleandcomparethattothepeaksintheU-234,U-235,andU-238,sotherewillbeaclearunderstandingofhowthelaboratoryisable to determine that U-233 is not present[emphasisadded,Reference1].”

TheEcologyDepartment’sefforttodeterminethatU-233isnotpresentintheHanfordReachriverbedcallsattentiontothestrengthofdifferentbeliefsinregardtoU–233.ManybelievethatU-233ofHanfordorigincertainlydoesnotcontaminateriverbedcolloids;itisamatterofidentifyingwhatevererrorshaveledtofalseclaimsforthepresenceofU-233.Forthem,severalmoreindependentstudieswillberequiredtoconvincethemtochangetheirbelief.

TRAC had proposed and tested a theory that U-233 of Hanford origin contaminates riverbed sediments.TRACestimatedthelikelihoodofU-233intheriverbedat80to95%,beforethepresentstudy.ForTRAC,thepresentstudywasanticipatedtobeanindependent,skepticalconfirmationofanalreadyestablishedtheorywithitssupportingevidence.

Asaconfirmationstudy,the18counts(sumofNcolumnintheAnalyticalResults)inthethreetestsamples plus duplicate are positive 4.2 standard deviations, in comparison to a total of 0 counts in the backgroundplustracerplusblank.Asaskepticalconfirmationstudy,theseresultsarepositive beyond reasonable statistical doubt.

�

Uranium-233 in the Hanford Reach Riverbed

Withthisindependent,skepticalconfirmation,beyondstatisticaldoubt,itwouldbejustasdifficulttoconvinceTRACthatU-233isnotpresentintheriverbedasitwouldbetoconvinceEcologythatU-233ispresent.

Thisgreatdifferenceinopinionhasarisenfromcorrespondinglygreatdifferencesinpre-existingbeliefs about the nature of contamination of the Hanford Site and of the Hanford Reach of the Columbia River.Toappreciatethepresentdifferenceofopinion,onemustrecognizehowstronglypresentinterpretationsofdatadependonpastexperiencesandfirmlyheldbeliefs.

Theissuesofwhatconstitutesasufficientdeterminationofscientificfactinthecontextofpre-existingbeliefsisthesubjectofBayesianstatistics.SeeReference7foraninformalintroductionto inferenceandtheinfluenceofevidenceonfirmlyheld,butopposingbeliefs.

Thedifferencebetweenthepreviouslyheld,stronglyconfirmingandstronglyrefutingbeliefswasincorporatedintothedesignofthereportedstudy.Advocatesofthetwodifferentbeliefsparticipatedinallphasesofthisstudy,undertheauspicesofthestateEcologyDepartment.Eachparticipantagreedwiththeprocedure.Therewasaninformalagreementtoaccepttheoutcomeofthisstudy.SamplesplitswereretainedbytheEcologyandtheHealthDepartmentsandTRACfortheirown,independentchecks,attheiroption.

Implications ThereadermightalreadyanticipatefromtheabovediscussionofConfidence,thatthereismuchatstakeovertheanswertothetechnicalquestionofwhetherU-233ofHanfordoriginispresentinsedimentsunderlying the Hanford Reach. Hundreds of thousands of dollars of labor and laboratory costs have been spentonthisquestionsince1999.

The presence of U-233 in riverbed colloids raises a specificconcernforthefuture health of biota in the riverbed of the Hanford Reach. The poster children of this concern are the hatchlings (alevin)ofwild-spawningfallchinooksalmonthatspendtheirfirstmonthsamongthecobblesintheHanfordReachriverbed.

Thisspecificconcernfocusesonradium-225(Ra-225),whichistheseconddecayproductofU-233.Radium-225mimics calcium. Calcium is an element essential for life. Salmon hatchlings are believed to absorb calcium —and hence Ra-225— fromriverbedwaterintheirhabitatundertheHanfordReach.

Radium-225hasahalf-lifeofonly15days.Thus,anyharmRa-225mightdothesalmonwouldoccuronlyduringthealevinstage.TheRa-225mightreducethestrengthofthewildsalmonstock,whichwouldaffectabillion-dollar-a-yearcommerce.

This study raises general concern for the degree that public and regulatory oversight of the Hanford Siteisweightedbypre-existingbeliefs.Reference2reportsunmonitoredradiumenteringtheriverinseepagesfromHanford’s300-Area.Thus,thepresentstudyjustifiesreviewofthedegreetowhichpre-existingbeliefsinfluencethemanagementstrategyandclean-upofthesite.

Animportantresultofthisstudyisthewiderangeofitsinterpretations:betweenpositiveconfirmationofU-233intheriverbedatoneextremetomereanalyticaldatarequiringfurtherprocedural

Salmon hatchling (alevin) - Photo: Army Engineer Corps.

�0

Uranium-233 in the Hanford Reach Riverbed

evaluationinthelaboratoryattheotherextreme.ThiswiderangerevealshowlittleisyetknownaboutHanfordSite’spresentandfutureimpactsontheColumbiaRiver.

A comprehensive, independent, environmental assessment of the Hanford Reach river corridor is neededtosupportdecision-makinginorderthatclean-upofHanfordSitewillprotectthepresentandthefuture of the Columbia River.

References(hyperlinkedwithinwww.radioactivist.org)

1•Hanford State of the Site Meeting Comments, Uranium-233 —letterbyJayManning,EcologyDirector(December2005)

2•Radioactive Bioaccumulation in Clams along the Hanford Reach (March2005)

3•Uranium-233 Detected in Hanford Reach Sediments (November 2003)

4•Trouble in the Columbia Riverbed —increasing radioactivity under the Hanford Reach (October 2003)

5•Hanford Radioactivity in Salmon Spawning Grounds —quality,extentandsomeimplications(August2002)

6•Questions and Answers on Uranium-233 at Hanford (August 2002)

7•Why Reasonable Minds Differ —learningmodeledonchangingconfidenceinbeliefs(April1997)

Produced and published by

The RadioActivist Campaign7312N.E.NorthShoreRd.Belfair,WA98528www.radioactivist.orgradioactivist@radioactivist.org

MoonCallison(layoutanddesign)mooncal@tscnet.com