Content uploaded by Rabiul Haque Biswas
Author content
All content in this area was uploaded by Rabiul Haque Biswas on Nov 17, 2014
Content may be subject to copyright.
MediterraneanArchaeologyandArchaeometry,Vol.2010,No.4,pp.9‐15
Copyright©2010MAA
PrintedinGreece.Allrightsreserved.
ASURVEYOFSOMENEWAPPROACHESINEXTENDING
THEMAXIMUMAGELIMITANDACCURACY
OFLUMINESCENCEAPPLICATION
TOARCHEOLOGICALCHRONOMETRY
A.K.Singhvi,N.ChauhanandR.H.Biswas
GeosciencesDivision,PhysicalResearchLaboratoryAhmedabad380009,India
Correspondingauthor:singhvi@prl.res.in
INTRODUCTION
Inrecentyearsconsiderableefforthasbeenmadetoextendthedatinglimitsandaccuraciesof
luminescencedatinganditsapplicationstoarcheologicalsciences.Theseinclude,forexample,the
useofsinglegrainsfordating(Duller2008:589‐612;JacobsandRoberts2007:210‐223),researchon
athermalfadingtoexploretheuseoffeldspars(withhighersaturationdose)fordating(Huntley
2006:1359‐1365;LamotheandAuclair1999:319‐323),theuseofredTLforthedatingofvolcanicash
(FattahiandStokes2003:647‐660;VisocekasandGuérin2006:942‐947)anddirectdatingof
archeologicalcontexts(ChawlaandSinghvi1989:416‐418;Singhvietal.1986:205‐207).Useofsome
ofthesehaveprovidedusefulnewdataonchronometryandhaveplacedluminescencedatingon
thecentrestageamongstotherchronometrictechniquesavailableforthedatingarcheologicalsites
andseveralimportantresultsonaspectsofhumandispersalandchronometryhavebeenreported
(Anikovichetal.2007:223‐226;Mellars2006:796‐800;Petragliaetal.2007:114‐116;Roberts1997:
819‐892;Singhvietal.1998:23-83).
Thiscontributionoutlinessomenewideasthatofferprospectsofdevelopingtheluminescence
datingtechniquefurtherandhelpextendtheagerangeofitsapplicationsinarcheology.These
applicationsarebasedonconsiderationofthemannerinwhichtheradiationdoseisdepositedin
naturalsamplesandaimtorefinetheprotocolsthatarecurrentlyusedfortheestimationof
equivalentdose.
KEYWORDS:redTL,volcanicash,fading,IRSL
A.K.SINGHVIetal
10
DATINGTHEINTERIOROFLITHICIM‐
PLEMENTS
Inluminescencedating,quartzhasbeen
usedwidelytoprovidetheburialagesofsedi‐
ments.Themaximumachievableagedepends
onthedosecorrespondingtothesaturationof
theluminescencesignalandthelifetimeofthe
stablesignal.Onthebasisoflifetimemeasure‐
mentsithasbeenestimatedthatthelumines‐
cencesignalofquartzisstablefor108ka.How‐
ever,onsetofsaturationoftheluminescence
signalatlowradiationdoses,limitsitsrangeof
applicability.Thesaturationofthelumines‐
cencesignalistypicallyoccursat~250Gy
(Chawlaetal.1998:53‐63).Thisrestrictsmaxi‐
mumageobtainedfromquartzgrainsto
aroundhundredka,anddependslargelyupon
theabsorbeddoseratefromthesedimentma‐
trix.
Inasedimentmatrix,theradiationenviron‐
mentcomprisesalpha,betaandgammaradia‐
tions,emittedduringthedecayofnaturallyoc‐
curringradioactiveisotopesofU,Thseries,K
and87Rb.Thealphaparticlerangeislimitedto
onlyfewmicrons,betahasarangeoffewmm
andgammadosehasarangeofabout30cm.
Underinfinitematrixassumption(Aitken1985),
typicallyfora100μmgrain,thegammadoseis
25‐30%ofthetotaldose.Inlargequartzgrains
thataregenerallydevoidofinternalradioactiv‐
ity,thedistributionofdoseinsidethegrain
fromtheexternalradiationenvironment
changeswithdepthduetoverydifferentranges
ofthealpha,betaandgammaradiations.Asa
simpleapproximation,thealphaandbetadose
getattenuatedwithdepthandtheircontribu‐
tionisnegligibleatdepthsfewmm.Thusfora
quartzitepebble,theinteriorofthegrainre‐
ceivedoseduetoonlyduetogammaradia‐
tions.Thissuggeststhatininteriorregion,the
doseratereducestoathirdorafourthofits
originalvalueand,inturnimpliesthatthesatu‐
rationoftheluminescencesignalinthisregion
wouldoccurovermuchlongertimescales.We
exploredthefeasibilityofusingtheinteriorof
largesizequartzgrainstoextendtheagerange
achievablebyluminescence(Chauhanetal.
2009:629‐633).Inthiswork,therangeofbeta
particlesfromdifferentradioisotopes(i.e.40K,
212Bi,214Biand234Pa)andtheirrelativedosecon‐
tributionwascomputedanalyticallyandbyus‐
ingMonteCarlosimulations.Fig.1showsthe
variationofdoseinsidethequartzgrainrelative
toinfinitematrixdoseinsedimentmatrix.
Fig.1Attenuationofbetadoseinsideaquartzgrain
duetovariousnaturallyoccurringradionuclides.The
radioactivityisconsideredbeexternaltothegrain.
Noticethatatdepthsof2mmmostofthebetadose
getsattenuated.
Thesecomputationscomprised,
1. estimationofbetadoseprofileinsidea
quartzgrainwithdepthnormaltosurface
usingMonteCarlosimulation,
2. developmentofanalyticalmethodologyto
computetheannualradiationdoseand
comparewiththesimulationresults,
3. estimationofthedepthatwhichbetadose
insidequartzgrainisreducedtoanegligible
value,and
4. examinationofthepossiblecausesthatmay
inhibittheapplicabilityofthenewmethods.
Resultsuggestthatinlargecmsizequartz
grains,thenetbetadose(fromallthepossible
betaemitters)isconfinedtoouter~2mmskin.
Theinteriorofquartzgrainreceivesradiation
doseonlyduetogammarays.Thisisabout30%
ofthetotaldoseratewhichimpliesthatbyana‐
lyzingtheinteriorofalargesizequartzgrain,
datingofasamplewhichisthreetofourtimes
oldershouldbepossible.
Basedonthis,wesuggestthepossibilityof
directlydatingheatedorsunexposed,archaeo‐
logicalartifacts.Microlithsandsmalllithicim‐
plementsaregoodcandidatestobeusedfor
datingbythisapproach,consideringthatoften
thesewereheatedtoahightemperatureto
ASURVEYOFSOMENEWAPPROACHESINEXTENDINGTHEMAXIMUMAGELIMITANDACCURACY
OFLUMINESCENCEAPPLICATIONTOARCHEOLOGICALCHRONOMETRY
11
makeiteasytoshapethem.Further,attimes
someoftheimplementsaretranslucentandthis
factfurtherensuresthattheirgeologicallumi‐
nescencecanhephotobleachedbydaylight
duringtheirmanufacture,use,disposaland
eventualburial.Initialexperimentswithlarge
(~cmsized)piecesofquartzindicatedthattheir
interiorregionswerealsowellbleachedpartly
duetothepresenceofmultiplecleaveplains
andfractures,thatensuredmultiplereflections
oflightinthelatticeandhencebleaching.The
extractionoftheinteriorofthegrainsisnon‐
trivialandtwopossibleapproachesare,1)Hy‐
drofluoricacidetchingforextendedperiods,or
2)theimagingofaslice(takennormaltothe
grainsurface)towardstheinteriorandusing
imagingtechniquestomeasurethespatialdis‐
tributionofdosethroughthesamplematrix.
Thiscanbedoneusingthespatiallyresolved
luminescencemeasurementsystemanalogous
tothatinitiatedbyGreilichetal.(2005:645‐665).
Asystemforspatiallyresolvedluminescenceis
beingconstructedatourlaboratoryandsuch
casesoftheinterioroflithicimplementswill
thenbeexaminedviathepaleodosegradientin
aslicethroughitsinterior.Also,itshouldin
principlebepossibletodateslithicimplements
withfiniteinternalradioactivityduetothefact
thatonlyapartofthebetadosearisingfromthe
internalradioactivityistrappedinsidethevol‐
umeconsidered.Inthiscontextwealsoreferto
anearlyworksbyLiritzis(1980:242‐251)and
HuntleyandPrescott(2001:687‐699),whoex‐
ploitedlowdoseenvironmentsfordatingover
anextendedtimerange.
SENSITIVITYCHANGESDURINGNATU‐
RALLUMINESCENCEMEASUREMENTS‐
THEIMPROVEDSARPROTOCOL
Generally,theSingleAliquotRegeneration
(SAR)protocol(MurrayandWintle2000:57‐73)
isusedtoestimatethedosedepositedina
sedimentmatrix.Inthisprotocol,thesensitivity
changeduringmeasurementismonitoredbya
testdose(TD)signal.Thenaturalandregenera‐
tionluminescencesignalsarenormalizedbythe
TDluminescencesignal.Thisprocedurethere‐
forecorrectsforsensitivitychangeduringa
dosemeasurementcycle.However,ithasoften
beenobservedthattheintensityofanatural
aliquotlieswayabovetheintensityofthere‐
generatedcurveandoftentheregeneratedin‐
tensityevensaturateswellbelowthenatural
intensity.Suchasituationcanonlyarisewhen
thesensitivities(luminescenceintensityperunit
massperunitdose)aredifferent,thatis,the
naturalluminescenceandtheregeneratedlu‐
minescencesignalsarerecordedwithdifferent
sensitivities.Itwasthereforepointedoutthata
finitechangeinsensitivitycanoccurduringthe
pre‐heatingmeasurementofnaturalOSLand
consequentlytheSARprotocolasusednow
doesnotfullycorrectforthesensitivity
changes.Laboratorystudiesindicatedthatsuch
sensitivitychangesduringOSLreadoutand
preheatcyclescanbesignificant(upto30%or
more)andcanthereforecausesignificantsys‐
tematicerrorsorevenmakeasampleunsuit‐
ableforanalysis(e.g.thecaseofnaturalsignal
exceedingthesaturationlimitoftheregenera‐
tionsignal)asshowninFig.2.
Fig.2Changesinthesensitivityof110°Cpeakduring
themeasurementcycles.Thelargestepsofchanges
areduetopreheats.TheSARprotocolcorrectsfor
theseeffectivelybuttheinitialchangeisnotincluded.
Theareaunderthepeakfrom90°Cto120°Cwas
integratedtoobtaintheintensityofthe110°Cpeak
Withthisproviso,werecommendadditional
measurementstepsfortheanalysisofasample
usingSARprotocol.Thiscorrectionprocedure
assumesthatthe110°CquartzOSLpeak
(MurrayandRoberts1998:503‐515)canbeused
asasurrogatefortheOSLsensitivityofthe
sample(acorrelationthatideallyshouldbees‐
tablishedforeachsample).Inthisprotocolthe
ratioofthe110°CTLpeakresponseforTD,re‐
A.K.SINGHVIetal
12
cordedbeforeandafterthemeasurementof
naturalOSLandthisratioisusedtocorrectfor
thesensitivitychangeduringOSLmeasure‐
ments(Fig.3).
Fig.3Theprotocolforsensitivitycorrectionduring
thepreheatandreadoutofthenaturalOSL.The
intensitiesofsteps2and5providethecorrection
factorNCF.
Fig.4ChangesinthedistributionofSARpaleodoses
withandwithoutNCFcorrection.Noticethereduc‐
tioninthespreadandtheshiftinthemeanvalue.
ThisratioistermedtheNaturalsensitivity
CorrectionFactor(NCF)andisusedtoensure
thatthenaturalOSLandtheregeneratedOSL
arecomparedwiththesamesensitivity.The
remainderoftheprotocolremainssimilarasfor
theSARprotocol.Finallytheequivalentdoseis
obtainedbysubtractingtheTDvalue(given
priortonaturalOSLmeasurement)fromthe
obtaineddosevalue.Fig.4givesthepaleodose
distributionofasampleanditisclearthata
noteworthyreductioninthespreadofpaleo‐
dosesoccurs,duepossiblytohithertounac‐
countedforchangesinthesensitivityinthe
conventionalSARprotocols.
Thisstudysuggests,arevisitoftheagesus‐
ingtheSARprotocolandasimpletestofa
comparisonofthesensitivityof110°Cglow
peaktoatestdosebeforeandaftertheOSL
readout,isrecommended.Recentexperiments
withdosedistributionfromavarietyofenvi‐
ronmentsshowsimilartrends.Adecreasein
scatterinthepaleodosesusingthisprotocol
wasseeninallthesamplestestedandthissug‐
gestedthatinmanysituations,wellbleached
samplesmaybemisunderstoodasapoorly
bleachedsample.Thisaspecthasbeendealt
withelsewhere(ChauhanandSinghvi2010:in
process).
DATINGOFVOLCANICASH
InIndianarcheology,thepresenceofarcheo‐
logicalimplementsincentral–westernIndia
andthedatingofassociatedvolcanicashlayer
havebeenextensivelydebatedonmethodologi‐
calgrounds.Thusforexample,Korisettaretal.
(1989:564‐566)firstdatedtheBoriashbyK‐Ar
methodto1.4Ma.Thiswaslaterrevisedto
600kausingAr–Arages.Ontheotherhand,
basedongeochemicalsimilarity,thislayerwas
consideredtobetheYoungestTobatuff(YTT)
datedto73±4kabyChesneretal.(1991:200‐
203).Recently,Petraglia,etal.(2007:114‐116)
datedapreandpostTobaarcheologicalsam‐
ple,inJwalapuram,AndhraPradesh,bylumi‐
nescencedatingtechniqueas77±6and74±7ka
respectively.Theagebracketingshowsthatthe
ageofthisashis~74Ka.Directluminescence
datingofTobaashwasfirstattemptedbyHorn
etal.(1993:326‐329).Thethermoluminescence
ageoftheglasses,derivedfromBoriAsh,was
23.4±2.4kaandwasinterpretedtodatethesec‐
ondaryreworkingandconsequentphotobleach‐
ingoflight‐sensitiveTLsignalsinthetuff,post
deposition.Analternativeexplanationwasthat
ASURVEYOFSOMENEWAPPROACHESINEXTENDINGTHEMAXIMUMAGELIMITANDACCURACY
OFLUMINESCENCEAPPLICATIONTOARCHEOLOGICALCHRONOMETRY
13
thisashsufferedathermalfading.Notmany
ashbedshavebeendatedintheIndiancon‐
texts.
Toresolvesomeofthebasicissuesofdating
volcanicashinIndia,weinitiatedasystematic
evaluationoftheashsampleusingavarietyof
approaches.Towardsthis,avolcanicashlayer
samplefromdifferentstratigraphiccontexts
wastakensuchthattheageoftheashbedcould
beconstrainedwithSARageonquartzgrains
fromhorizonsthatsandwichtheashlayer.
Basedonfieldevidences,oneoftheashlayers
wasalsoconsideredastheTobaash.Datingof
sediments,aboveandbelowtheashlayer,us‐
ingSARprotocolforquartz,gaveagesof~52
and~76kasuggestingtheageoftheashlayer
tobewithintheagerangeof52‐76ka.These
ageshoweverdidnotpermitaclearresolution
ontheageexceptthatthedifferencebetween
thisandagesonhorizonsaboveandbelow.Di‐
rectdatingoftheashusingTL‐ MultipleAli‐
quotAdditiveDosemethod,inblueemission
window(430±30nm)yieldedanageof47±8ka.
ThisageistwicethatobtainedbyHorn,etal.
(1993:326‐329),suggestingthattheirinference
ofashbedbeingreworked/redeposited(and
hencenotinprimarycontext)waspossiblycor‐
rect.Usingacorrectionprocedurereportedby
Someshwararao(1996),thesensitivitycorrec‐
tionprocedureforathermalfadingindicateda
correctionfactorof1.6fortheunderestimated
ages.TheprotocolisoutlinedinFig.5.
Fig.5Correctionprocedureforfadingusingasample‐
basedanalysisfollowingSomeshwarararao,1996.
Thecorrectionfactoristheratioofsensitiv‐
ityatthetimeofburial(S2=S21‐S22)tothatas‐
receivedsample(S1=S11‐S12).Thisresultedina
valueof75±10kaandisclosertoitsbeingthe
YTTash.WethenattemptedthePostIRIRSL
SARprotocol,asintroducedbyBuylaertetal.
(2009:560‐565)(Fig6)andthepreliminaryre‐
sults,gaveanage~70ka.
Fig.6PostIRIRSLSARprotocolforthedatingofvol‐
canicash(followingBayleaurtetal.2009).
ItthusseemsthattheboththeSensitivity
correctionandthepostIRIRSLSARprotocol
workwell.Fig.7providestypicalopticaldecay
curvesobtainedusingIRSLandpostIRIRSL
elevatedtemperaturemeasurements.Theques‐
tionremainsthatastowhytheyprovidesimilar
resultsandthephysicsthatleadstothiscon‐
vergenceneedsfurtherelucidation.
Fig.7AtypicalIRSLandpostIRIRSLopticaldecay
curveforafinegrainsampleofvolcanicash.
A.K.SINGHVIetal
14
CONCLUSIONS
Theconceptualformalismofthreemethodo‐
logicalaspectsofluminescencedatinghasbeen
presented.Feasibilitystudiesonthese,
1. showthepossibilityofincreasingthe
datingrange,threefold,
2. indicatethepotentialfordecreasingthe
systematicerrorsthatcanoccurinSAR
protocolinOSL,and
3. providetheprospectsofimprovingthe
accuracyofagesinthecaseofvolcanic
ashbeds.Wesincerelyhopethatthisre‐
portwillinduceothergroupstotesttheir
efficacy.
ACKNOWLEDGEMENTS
AKSthanksS.StokesforearlycollaborationonthesensitivitycorrectionproceduresinSARpro‐
tocol.WethankDr.R.RajandDr.N.JuyalforhelpwithsamplecollectionsandMr.P.Adhayaru
forhissustainedhelpwiththeinstrumentation.Wethankstherefereesfortheirpainstakingcom‐
mentsandforpointingoutlanguageflaws.Butforthem,themanuscriptwouldnothavebeenas
palatable.
REFERENCES
Aitken,M.J.1985,Thermoluminescencedating.London:AcademicPress
Anikovich,M.V.,A.A.Sinitsyn,J.F.Hoffecker,V.T.Holliday,V.V.Popov,S.N.Lisitsyn,S.L.For‐
man,G.M.Levkovskaya,G.A.Pospelova,I.E.Kuzʹmina,N.D.Burova,P.Goldberg,R.I.
Macphail,B.GiaccioandN.D.Praslov2007,EarlyUpperPaleolithicineasternEuropeand
implicationsforthedispersalofmodernhumans.Science315:223‐226.
Buylaert,J.P.,A.S.Murray,K.J.ThomsenandM.Jain2009,Testingthepotentialofanelevatedtem‐
peratureIRSLsignalfromK‐feldspar.RadiationMeasurements44:560‐565.
Chauhan,N.,S.Anand,T.PalaniSelvam,Y.S.MayyaandA.K.Singhvi2009,Extendingthemaxi‐
mumageachievableintheluminescencedatingofsedimentsusinglargequartzgrains:A
feasibilitystudy.RadiationMeasurements44:629‐633.
Chauhan,N.andA.Singhvi2010,DispersioninSARPaleodosesduetoSpatialHeterogenietyof
NaturalBetaDose.Geochronometria(submitted):
Chawla,S.,T.K.GunduRaoandA.K.Singhvi1998,Quartzthermoluminescence:Doseanddose‐
rateeffectsandtheirimplications.RadiationMeasurements29:53‐63.
Chawla,S.andA.K.Singhvi1989,Thermoluminescencedatingofarchaeologicalsediments.Natur‐
wissenschaften76:416‐418.
Chesner,C.A.,W.I.Rose,A.Deino,R.DrakeandJ.A.Westgate1991,EruptivehistoryofEarthʹs
largestQuaternarycaldera(Toba,Indonesia)clarified.Geology19:200‐203.
Duller,G.A.T.2008,Single‐grainopticaldatingofQuaternarysediments:Whyaliquotsizematters
inluminescencedating.Boreas37:589‐612.
Fattahi,M.andS.Stokes2003,Redluminescencefrompotassiumfeldsparfordatingapplications:
Astudyofsomepropertiesrelevantfordating.RadiationMeasurements37:647‐660.
Greilich,S.,U.A.GlasmacherandG.A.Wagner2005,Opticaldatingofgraniticstonesurfaces.Ar‐
chaeometry47:645‐665.
Horn,P.,D.Müller‐Sohnius,D.StorzerandL.Zöller1993,K‐Ar‐,Fission‐Track‐,andThermolumi‐
nescenceAgesofQuaternaryVolcanicTuffsandtheirbearingonAcheulianArtifactsfrom
Bori,KukdiValley,PuneDistrict,India.ZeitschriftderDeutschenGeologischenGesellschaft
144:326‐329.
Huntley,D.J.2006,Anexplanationofthepower‐lawdecayofluminescence.JournalofPhysicsCon‐
densedMatter18:1359‐1365.
ASURVEYOFSOMENEWAPPROACHESINEXTENDINGTHEMAXIMUMAGELIMITANDACCURACY
OFLUMINESCENCEAPPLICATIONTOARCHEOLOGICALCHRONOMETRY
15
Huntley,D.J.andJ.R.Prescott2001,Improvedmethodologyandnewthermoluminescenceagesfor
thedunesequenceinsouth‐eastSouthAustralia.QuaternaltyScienceReviews,20:687‐
699.
Jacobs,Z.andR.G.Roberts2007,Advancesinopticallystimulatedluminescencedatingofindivid‐
ualgrainsofquartzfromarcheologicaldeposits.EvolutionaryAnthropology16:210‐223.
Korisettar,R.,T.R.Venkatesan,S.Mishra,S.N.Rajaguru,B.L.K.Somayajulu,S.K.Tandon,V.D.
Gogte,R.K.GanjooandV.S.Kale1989,Discoveryofatephrabedinthequaternaryallu‐
vialsedimentofPunedistrict(Maharashtra)PeninsularIndia.CurrentScience58:564‐566.
Lamothe,M.andM.Auclair1999,Asolutiontoanomalousfadingandageshortfallsinopticaldat‐
ingoffeldsparminerals.EarthandPlanetaryScienceLetters171:319‐323.
Liritzis,Y.1980,PotentialuseofThermoluminescenceindatingspeleothems.Anthropos 7:242‐251.
Mellars,P.2006,Goingeast:Newgeneticandarchaeologicalperspectivesonthemodernhuman
colonizationofEurasia.Science313:796‐800.
Murray,A.S.andR.G.Roberts1998,Measurementoftheequivalentdoseinquartzusingaregen‐
erative‐dosesingle‐aliquotprotocol.RadiationMeasurements29:503‐515.
Murray,A.S.andA.G.Wintle2000,Luminescencedatingofquartzusinganimprovedsingle‐
aliquotregenerative‐doseprotocol.RadiationMeasurements32:57‐73.
Petraglia,M.,R.Korisettar,N.Boivin,C.Clarkson,P.Ditchfield,S.Jones,J.Koshy,M.M.Lahr,C.
Oppenheimer,D.Pyle,R.Roberts,J.L.Schwenninger,L.ArnoldandK.White2007,Mid‐
dlepaleolithicassemblagesfromtheIndiansubcontinentbeforeandaftertheTobasuper‐
eruption.Science317:114‐116.
Roberts,R.G.1997,Luminescencedatinginarchaeology:Fromoriginstooptical.RadiationMeas‐
urements27:819‐892.
Singhvi,A.K.,W.Sauer,G.A.WagnerandK.Kroger1986,Thermoluminescencedatingofloessde‐
positsatplaidterhummerichanditsimplicationsforthechronologyofNeanderthalman.
Naturwissenschaften73:205‐207.
Someshwararao,M.1996,Studiesonphysicalbasisofluminescencegeochronologyanditsapplica‐
tions.UnpublishedPh.D.Thesis154pp.
Visocekas,R.andG.Guérin2006,TLdatingoffeldsparsusingtheirfar‐redemissiontodealwith
anomalousfading.RadiationMeasurements41:942‐947.