1

⁰

AtypicalBeadCulturedPinctadamaximaPearls

NucleatedwithFreshwaterNon-BeadCulturedPearls

PromlikitKessrapongandKwanreunLawanwong

Figure1.The12whiteand“golden”SouthSeaculturedpearls(Pinctadamaxima)ofvariousshapesandsizesexaminedinthisstudy.Althoughtheirexternalappearancesaresimilartosaltwaterbeadandnon-beadcultured(NBC)pearls,11ofthemareatypicalbeadcultured(aBC)pearlsformedbyinsertingfreshwaternon-beadculturedpearlsasnucleiintosaltwaterhosts.PhotobyNuttapolKitdee.

NEWSFROMRESEARCH

2

⁰

AbstractAtpresent,mostofthefreshwater (FW) shellbeadnucleiusedintheculturedpearl industryoriginatesfrommusselsfishedintheMississippiRiver,UnitedStates (Strack,2006;Hsuetal.,2016).However,shellbeadsarenottheonlymaterialusedasnuclei intheculturingprocess. Cultured freshwater pearls have also been used as nuclei to produce atypical bead-cultured(aBC) SouthSeapearls.Twelve aBCpearlsamples fromOrientPearl (Bangkok)Limitedwereexaminedwithstandardpearltestingmethods. MicroradiographyandX-rayluminescenceexaminationwereappliedtoall thesamples, whileRamanand photoluminescence(PL) spectroscopyaswellas ultravioletvisiblespectrophotometry(UV-Vis)analyseswerecarriedoutonthoseofa“golden”hue. Twoofthesamples,onewhiteandone“golden,” wereselectedinordertocut inhalf andcollectfurtherdata. ThesawingenabledtheX-rayluminescenceoftheinneroverlyingsaltwater (SW) layers and freshwater pearl nuclei to be recorded and provided the opportunity to takeadditional photomicrographs and perform LA-ICP-MS chemical analysis across the innersurfacesofonehalfofeachpearl.Asexpected,theresultsonthetwocutsamplesinparticularconfirmedtheirstatedaBCnature. TheinternalstructuresrevealedclearboundariesbetweenthefreshwaterNBCpearlnucleiandthe saltwater overgrowth layers. Raman, PL, and UV-Vis spectra for the “golden” sampleindicated the pearl formed within a Pinctada maximamollusk. Also, in keeping with theirfreshwaterNBCnucleistructure,aweaktomoderatefluorescencewasobservedintheintactsamples,whichwould not be likely if theywere entirely saltwater. The cut examples clearlyshowedthatthestrongfluorescenceoriginatedfromthenuclei,asisthecasewithtraditional(typical) beadculturedmarinepearlsinwhichfreshwatershellnucleiareimplanted. LA-ICP-MSanalysisonthetwocutsamplesclearlyrevealedthenatureof theoutersaltwater layersandinner freshwater pearls, thus confirming the fluorescence reactions and allowing clearidentificationofthechemicalelementswithinthetwocomponents. Standard pearl testingmethods permit the straightforward identification of such aBC pearls. Cuttingthepearlsinhalfisausefulmethod tocarryoutfurther scientificexaminationinordertoobtain moredatanotpossiblefromtheintactpearls.

3

⁰

TableofContentsAbstract…………………………………………………………………………………………………………………………………….2

1.Introduction………………………………………………………………………………………….……………………………….4

2.Materialsandmethods……….……………………………………………………………………………………..………….4

3.Resultsanddiscussion………………………………………………….………………………………………………………..6

3.1Real-timeMicroradiography……………………………………………………………………………………………6

3.2X-rayComputedMicrotomography(µ-CT)…………………………………………………………………….11

3.3OpticalX-rayFluorescence………………………………………………………………………………………….…14

3.4RamanandPLSpectroscopy……………………………………………………………………………………….…16

3.5VisibleSpectrophotometry…………………………………………………………………………………………….18

3.6LA-ICP-MS………………………………………………………………………………………………………………….…..19

4.Conclusion……………………………………………………………………………………………………………………………23

5.AbouttheAuthors……………………………………………………………………………………………………..………..24

6.Acknowledgments………………………………………………………………………………………………………………..24

References……………………………………………………………………………………………………………………………25

4

⁰

1.IntroductionWhiteandyellow“golden” (GemologicalInstituteofAmerica,2011) SouthSeaculturedpearlsare produced by thePinctadamaximamollusk,which inhabits thewaters of the Indian andPacificOceans (Strack,2006).The leadingsourcesofsuchpearlsareAustralia, Indonesia, the Philippines,andMyanmar. Duringthetraditionalbeadculturingprocess,freshwatershellbeadnuclei from North America are routinely inserted intomollusks. In this study, atypical beadculturedpearls inwhichfreshwaternon-beadcultured(NBC) pearlswerereportedly insertedas nuclei are examined and described (figure 1). Most of the shell bead nuclei used in thecultured pearl industry originates from freshwatermussels (Sturman et al., 2016). However,shell beads are not the onlymaterial used as nuclei in the culturing process (Scarratt et al.,2017). Asthisworkshows,culturedfreshwaterpearlshavealsobeenusedasnucleitoproduceatypicalbeadcultured (aBC) SouthSeapearls. Aswouldbeexpected,theexternalappearancesof such pearls match cultured pearls produced by Pinctada maxima mollusks. However,misunderstandingsmayeasilyariseconcerningtheexactnatureofthepearls,andhencetheirperceived value, if their true identity is not revealed by advanced testing. This work clearlyillustratestheinternalcharacteristicsofthisparticulartypeofatypicalbeadculturedpearl.2.MaterialsandmethodsIn2017,12 South Sea culturedpearl samples thathadbeenproducedusing freshwaterNBCpearlnucleiwereobtainedfromOrientPearl (Bangkok)Ltd. Theywereselected inpersonbyGIA Thailand pearl researcherswho had the opportunity to visit the company’s factory. Thesamples loaned were semi-baroque to baroque of white, light yellow, and strong yellow tostrongorangyyellow colors. Theyweighedfrom4.47to11.45caratsand ranged from11.00×9.93×6.89mmto18.27×11.17×9.62mm.

Figure2. Twosampleswereselectedformoredetailedwork. Eachpearlwassawninhalfat GIA’sfieldgemologydepartmentinBangkoktostudythe internalstructure. PearlA (left)isorangyyellowandmeasures 18.27×11.17×9.62mm,andpearlB (right) iswhiteandmeasures 17.07×10.25×8.56mm. PhotosbyPromlikitKessrapong.

A B

5

⁰

Theinternalstructuresofallthesampleswereexaminedwith a PacificX-rayImaging(PXI) GenX-90Pseriesreal-timeX-ray(RTX) unitwitha4-micronmicrofocus,90kVvoltageand0.16 mAcurrentX-raysource,incorporatingaPerkinElmer4"/2" dual-viewflatpaneldetector. AProConCT-miniX-raysystemwitha5-micronmicrofocus,90kVvoltageand0.18mAX-raycurrentsource,andaVareximagingflatpaneldetectorwith74.8μmpixelpitchwereusedtoobtainX-raycomputedmicrotomographydatafromthreeselected samplesafterreviewingtheRTXresults.A FocalSpot Verifier PF-100 model FSX-PF100 X-ray fluorescence unit (voltage 100 kV andcurrent 3.2mA) was then used to provide information about the environment inwhich thepearlandnucleiformed,sincemarineandfreshwatermaterialsreactdifferentlywhenexposedtoX-rays(Hännietal.,2005).

Raman,PL,andUV-Visspectroscopy analyseswere carriedoutontheseven“golden”samplesinthegroup(figure1). An Agilent Cary60 UV-Visspectrophotometerutilizingacustom-madeGIA integratedsphereaccessory incorporatingan80Hzxenonflash lampsourcewasusedtocollectspectrainthe200–800nmrange. ARenishawinViaReflexmicro-Ramanspectrometersystemequippedwith a514nmargon-ionlaserwasusedto obtainRamanandPLspectrafromthe external and internal surfaces. The laser was set at 100% power utilizing threeaccumulationsandanexposuretimeof10seconds. PLspectrawerecollectedwiththelaser setat 50%powerandutilizingoneaccumulationwithanexposuretimeof15seconds. The two samples selected for sawing (figure 2) were cut in half using a Lapcraft Dia-Laserdiamond saw to reveal their internal structures. The surfaceswere groundprior to polishingusinga600#diamondlapandthenpolishedwithaSmartCutUKAMIndustrialSuperhardToolslapandLightSidegrit. PhotomicrographyoftheinteriorfeaturesofthetwocutsampleswasperformedusingaNikonSMZ18microscopeincorporatingNIS-Elementsimagingsoftware.Traceelement chemistry resultswereobtainedononehalf ofeachsawnsampleusing aThermoFisherScientificiCAPQc inductivelycoupledplasma–massspectrometer(ICP-MS) coupledwithaQ-switchedNd:YAGlaserablation(LA) deviceoperatingatawavelengthof213nm. ICP-MSwasoperatedusingtheforwardpowerat~1350Wandthetypicalnebulizergasflowat~0.80L/min. Heliumwasusedasthecarriergasinthelaserablationunit,andtheflowratewassetat~0.80L/min. Laserconditionswere40mmdiameterlaserspots,afluenceofaround10J/cm2,anda10Hzrepetitionrate. LA-ICP-MSanalysisisaquasi-nondestructivemethodthatcreatessmalllaserablationspotsthroughoutthetestingarea(Homkrajaeetal.,2019).

6

⁰

3.Resultsanddiscussion3.1Real-timeMicroradiographyReal-timemicroradiographsofall12sampleswereobtained,andsomeofthoserelatedtothe10 intact pearls are shown in table 1. The results for the remaining two pearls that weresubsequently selected for sawing (pearls A and B) and described in a preliminary work(KessrapongandLawanwong,2018) areshownintable2, togetherwithphotomicrographsintheirsawnstate. AllthesampleswereX-rayedinat leastthreedirections,twoofwhichhave beenselectedtoshowtheinternalstructureofeachsampleintable 1. Whenthesampleswereanalyzedside-on(table1,left) theytendedto revealthe clearestdetailswith thedemarcationbetweenthenucleusandoverlyingnacrebeingmostobvious in thisorientation.An irregulardemarcationoutlinebetweenthenucleusandsurroundingnacre,noobviousgrowthfeaturesin the overlying nacre, and twisted linear-like features typical of FW pearls (Scarratt et al.,2017) was shown in all but one of the samples. Experienced pearl testers may be able toaccurately identify the environment inwhich a pearl formed from studying the internal RTXstructures, especially NBC pearls, and as a result make a confident determination on theidentity of a pearl as well. When, as in the majority of these pearls, there is an apparentmismatch of structures from one environment with the appearance and expected origin ofanother environment, concern about the true identity arises. The concerns become morepronouncedwhenanirregulardemarcationboundaryappears inmostofthesamples. Ashasbeen discussed in the literature already, such demarcation boundaries are diagnostic of aBC pearls inwhichavarietyofnon-shellmaterialshavebeenusedasnuclei (Hänni etal.,2010;Scarrattetal.,2017).

However, the large void and organic-rich structure of pearl 6 (table 1) stands out from thestructures oftheothers. Sincethere isnoobviousdemarcationboundary, it isclear that thispearl is either the odd one out in the group or, more likely, an NBC pearl (Sturman, 2009;Sturmanetal.,2016;Nilpetployetal.,2018).

7

⁰

Table1.Macroimages(left)oftenpearls,side-onorientedmicroradiographyimages(middle),and“facingview”(thinnestdirection)orientedmicroradiographyimages(right).

PearlNo.

Macroimage Microradiographyimages

Sideview Facingview

1

2

3

4

5

8

⁰

6

7

8

9

10

9

⁰

PearlsAandBweresawninhalftoconfirmtheiridentificationandobserve/recordtheresultsfromfurtherworkthatcouldnotbeperformedontheintactpearls. Theinnerfeaturesofsawnsurfacesareshowninthephotomicrographsintable2. PearlAshowedadistinctboundarybetween the white pearl nucleus and the yellow nacre overgrowth, while the boundarybetweentheaBCnucleusandovergrowthlayersinpearlBwasmuchhardertodefineowingtothesimilarityincolorsbetweenbothcomponents. Bothpearlsalsoshowedfeaturestoonesideofthenucleusthataretypicalofadditionalgrowthformsencounteredinsomepearls. Theseareoftenmanifestedastwin-typepearlswithtwopointsoforiginthatcombinetoformasingleentity. Insomecases,pearls(natural,beadcultured,orNBC) withmultipleparts,allwiththeirowncore,maycombinetoformonebaroque pearl. RTX images(table2) showtheinternalstructuresinrelationtotheorientationofthephotomicrographsshown.AlongwithanobviousnucleusdemarcationboundaryinpearlA,acomplexlinear-likestructurewasalsovisibleinthecenterofthenucleusandanadditionalelongatedstructure (yellowarrows,table2)wasapparentinthetwinpartofthepearl. Furthermore,a“seed-like”satellitefeature (redarrow,table2) oftenassociatedwithsaltwaterculturedpearls (Krzemnicki,2010;Sturmanetal., 2015; Nilpetployetal.,2018) wasalsoobserved. PearlBrevealedaboundary/fold (whitearrow,table2)separatingthepearlintwoparts. ThelargestpartshowedanobviousFWaBCnucleuswithacomplex linear feature similar to that seen in pearl A, while the smaller part contained anirregularvoidtypicalofSouthSeaNBCpearls.

10

⁰

Table2. Microphotographyimages (left) ofoneofthehalvesofeachsampleandthecorrespondingorientedmicroradiographyimage (right). PhotosbyKwanreunLawanwong.

Sample Microphotography Microradiography

PearlA

PearlB

11

⁰

3.2X-rayComputedMicrotomography(µ-CT)

Eventhoughthesamplesrevealedfairlydefinedatypicalbeaddemarcations,someofthemwerenotquiteaspronouncedandsoitwasdecidedtoperformμ-CTanalysistobetterdiscerntheirinternalstructures. Three samples, pearls 3,6, and9(table1),wereselectedbecausetheirstructureswarrantedfurther investigation. The μ-CT results in table 3 revealed more defined irregular nucleusoutlines and complex linear features characteristic of FW NBC pearls in the center of eachnucleus in pearls 3 and 9, while the absence of any clear growth features in the layersovergrowing the nucleus was also noted. The complex central linear feature in pearl 3 alsocontained some tiny white spots within it and also in the gap between the boundary (redarrows,table3andfigures3and4). Althoughsuchseedsarenotdiagnosticofaculturedorigin,the authors’ experience and the previous studies (Krzemnicki, 2011; Nilpetploy et al., 2018)haveshownthistobemoreindicativeofculturedpearlformation.Likewise,thecomplexlinearfeatureinthecenterofpearl9’snucleus,especiallyintheXandZ directions,is characteristicofFWNBC pearls (Krzemnicki, 2010), thus supporting the conclusion that a FWNBC pearlwasusedasa“bead”nucleus.

AswassuspectedfromtheRTXevidence,pearl6didnot revealanobviousdemarcationlinethatwouldprovethepresenceofasimilarFWNBC“bead”seeninalltheothersamples. Basedonthefeaturesevident,especiallythelargevoidwithorganic-richmatter,thissampleisanNBCpearl thatwasmistakenlyincludedwithinthegroup. Thisshowsjusthoweasyitistomixpearlsofdifferentidentitiesbasedsolelyontheirexternalappearance.

12

⁰

Table3. Theμ-CTslicesofthethreepearlsamples inthreedifferentdirections.

PearlNo.

μ-CTimages

X-direction Y-direction Z-direction

3

6

9

13

⁰

Figure3.Amagnifiedμ-CTslicefromtheY-directionofpearl3(table3)showingawhiteseed-likestructureinthecentralirregularvoidfeature(seeredarrow).

14

⁰

Figure4.Amagnifiedμ-CTslicefromtheY-directionofpearl3(table3)showingawhiteseed-likestructureintheorganic-rich demarcation area between the FW NBC pearl nucleus and the overlying SW pearl nacre (see redarrow).

3.3OpticalX-rayFluorescenceTheuseofthisanalyticaltechniqueinpearltestinghasbeencoveredintheliterature(Hännietal.,2005;Kessrapongetal.,2017;Karampelasetal.,2019). However,ingeneralthereactionsofpearlstoX-rayexcitationfallintotwogroups; thosethatdonotreact(inert) andthosethatdoreact. The latter may be subdivided into two groups, the first showing very weak to weakreactions—someNBCandmost saltwaterBCpearlswith a FWshell beadnucleuswhere thenacreisnottoothick—andthesecondwherethereactionisusuallyverystrongandindicatesafreshwaterorigin. Theintensityofapearl’s fluorescence reaction isusuallydirectlycorrelatedto the levelof manganese (Mn)within it. SinceFWpearlsusuallycontainmoreMnthanSWpearls, it is no surprise to see that FW pearls often show a stronger reaction. On the otherhand, pearls that form in a saltwater environment are generally inert. As inmost scenarios

15

⁰

thereareexceptionstotheserules,anditispossibletoencounterbothFWpearlswithlowerthanusual(<100ppmw) Mnlevelsthatdonotfluorescestronglyandrareexamplesofmixed-chemistrypearls(Sturmanetal.,2019) thatshowreactionssimilartothosenormallyobservedinFWpearls. ElevenofthepearlsinthegroupexhibitedveryweaktomoderateX-rayfluorescencereactions,andonlypearl6wasinert. Unsurprisingly,theyellowandorangyyellowpearlsshowedtheweakestreactions. Thiswasduetoacombinationofthemaskingeffectoftheircolor,thetendencyofwhitetocreampearlstoshowstrongerreactions,andthenacrethicknessofthelayerssurroundingtheFWnucleusineach. Pearl6 wasinertbecauseitisanNBCpearlandthuslackstheFWnucleusobservedintheremainingsamples.Thetwopearlsselectedforsawingrevealeddistinctreactions thatwerenotsoapparentintheirintactforms. TheFWNBCpearlnucleusineachproducedweaktomoderatereactionsthroughtheSWnacrelayersovergrowingthem,withpearlAdisplayingyellow/orange colorandpearlBdisplayingayellowishgreencolor (figure5B). Thereactionsof thesawnpieces were much more revealing and showed striking differences between the FW and SWcomponents(figure5D). X-rayfluorescenceisanextremelysensitiveandusefultestavailabletogemologists. Theresultsobtainedfrombothsawnpearlsshowedthatthenucleus inpearlAfluorescedayellowcolorsimilartothatseenexternallypriortosawing,andthatthenucleusofpearlBfluoresced toasimilardegree,onlywithagreenishhue. TheseresultssupportedtheidentityobtainedfromtheRTXworkandhelpedconfirmthattheywereaBCpearlswithaFWNBCpearlusedasthenucleus.

16

⁰

Figure5. Imagesofthetwopearlsamples selected,showingthemas wholepearls(A) andtheirreactionsunder opticalX-rayfluorescence(B). Theinnersurfaces oftwohalves ofeachofthepearls areshown inLEDlighting(C)andunderX-rayexcitation(D).

3.4RamanandPLSpectroscopySevenpearlsfromthegroupexhibitedayellowtostrongyellowcoloration,andeachofthemwasalsoexaminedwithRamanandPLspectroscopytoconfirmtheircolororigin. Allrevealedspectra consistent with naturally colored Pinctada maxima (gold-lipped) shell and pearlsstudiedpreviously(Elen,2001;Zhouetal.,2012;CartierandKrzemnicki,2016). Twodifferentareas,theoutersurfaceandsimilarlycoloredlayersovergrowingthenucleusofsawnpearlA, wereexaminedtocomparetheresults. TheRamanspectrashowedaragonitepeaksat701/705(doublet), 1085, and 1462 cm–1 (Urmos et al., 1991), aswell as pigment-related features ofvariousintensitiesatapproximately1388and1542cm–1 (CartierandKrzemnicki,2016) relatedtothecolorsaturationwithintheareastested. Thiscanclearlybeseeninfigure6,wherethespectrum obtained from the inner area shows more pronounced features at approximately1388 and 1542 cm–1 than those in the spectrum obtained from the outer surface. The PLspectra of all samples, includingpearl A, exhibited a domed form typical of previous studies (Zhouet al., 2012;Kwaket al., 2016)withamaximumat around650nm. Nootherobviousfeatures were noted, save one at 544 nm associated with the dominant aragonite peak

A B

C D

PearlA

PearlB

PearlA PearlB

17

⁰

observedat1085cm–1intheRamanspectra. InkeepingwiththeRamanspectra,thePLresultfrom the inner surface of pearl A showed higher fluorescence,while the spectrum from theoutersurfacerevealedmuchweakerfluorescence (figure7). Thedifferencesintheintensitiesandspectral features seenmaybeexplainedby themoresaturatedcolorof the inner layerscomparedtothelighter-colored outernacre.

Figure6. RamanspectraofpearlA,outersurface(blueline) andinnersurface(redline),showingaragonitepeaksat701/705,1085, and1462 cm–1. Thepigment-related featuresat1388and1542cm–1 are clearly visible in thespectrumcollectedfromthemoresaturatedcolorlayerssurroundingthenucleuswithinthepearl.

0

20000

40000

60000

80000

100000

120000

140000

160000

100 300 500 700 900 1100 1300 1500

RamanSpectraofPearlA

Inner surface

Outer surface

RAMANSHIFT(cm–1)

INTENSITY(C

OUNTS)

1085

1542 1388

701/705

1462

18

⁰

Figure7. Photoluminescence spectraofpearlA,outersurface(blueline) andinnersurface(redline),showingdomedspectrawithamaximumatapproximately650nm. Thefeatureat544nmisassociatedwiththearagonitepeakat1085cm–1intheRaman (figure6). Notethehigherfluorescenceexhibitedbytheinnersurface.3.5VisibleSpectrophotometryTheUV-Visspectraobtainedfromtheexternalsurfacesoftheseven“golden”samplesrevealedan area of lower reflectance between approximately 330 and 460 nm, higher reflectancebetweenapproximately480and700nm,andahigh-reflectancefeatureatapproximately310nm,togetherwithalow-reflectancefeatureat280nm (figure8). Allthefeaturesareconsistentwith thoseobserved inknownsamplesofnaturally coloredyellowPinctadamaxima samplesthe authors have examined in GIA’s database and that have also been recorded in prior literature(Elen,2001;CartierandKrzemnicki,2016). Theauthorswereunabletoobtainspectraofhighenoughqualityfromtheinnersurfacesofbothsawnsamples toreproducehere.

0

5000

10000

15000

20000

25000

517 567 617 667 717 767 817

PLSpectraofPearlA

Inner surface

Outer surface

RAMANSHIFT(cm-1)

INTENSITY(C

OUNTS)

650

544

19

⁰

Figure8. TheUV-Visreflectancespectra offourrepresentativepearls(A,2,3,and10) showing lower-reflectancefeaturesat330–460nmaswellasahigher-reflectancefeatureat310nm. Pearl10exhibitedthelightestcolorandpearl2themostsaturatedcoloration.3.6LA-ICP-MSPearlsfromFWandSWenvironmentsareseparatednotonlybytheopticalX-rayfluorescencemethod described previously. The quantities of the trace elements—especially manganese(Mn),strontium(Sr), andbarium(Ba)—presentwithinshellsandpearlsnecessarytomaketheseparationmaybedetectedusingenergydispersiveX-ray (EDXRF) spectrometryand/or laserablation–inductively coupled plasma–mass spectrometry (LA-ICP-MS). The latter method ofanalysis is useful when more detailed results are needed, though the analysis spot is onlyaround 40 microns across and in depth (Abduriyim, 2018; Karampelas et al., 2019).HigherquantitiesofSrindicateapearlformedinasaltwaterenvironment,whilehigherquantitiesofMnindicateafreshwaterenvironment.ThemajorityofsaltwaterpearlstestedinGIAlaboratoriespossessSrlevelsbetween1,000and2,000ppmw,whileinfreshwaterpearlsthequantities mayfallbelowdetectionlimits(BDL)to200 ppmw atmost. A higherMn content is themain indicator of formation in a freshwaterenvironment,withtheusualrangefrom100to5,000 ppmw. TheEDXRFresultsontheexternal

0

20

40

60

80

100

120

140

160

200 300 400 500 600 700 800

PearlA

Pearl2

Pearl3

Pearl10

WAVELENGTH(nm)

REFLEC

TANCE

(%)

310

280

UV-VisReflectanceSpectra

20

⁰

surfacesofthepearlsstudiedprovedtheyoriginatedfromaSWenvironment. However,sincethevariousX-ray-based techniquesappliedprior to theLA-ICP-MSanalysisproved thataFWNBC pearl nucleus existed in 11 of the samples, the decision was taken to cut two of thesamplesinhalftostudytheirinnerchemistryresultsandfluorescencereactions. Thefindingsofthe LA-ICP-MS analysis on one half from each pearl (figures 9 and 10) proved that theNBCnucleiusedwereFWinnaturewhiletheovergrowth layerswereSW,pertheexteriorEDXRFanalysisoriginallyundertaken.Atotalof58laserablationspotswerepositionedacrosstheinnersurfaceofpearlA(figure9) inalinefromtheexteriorsurfaceofoneside(spot#1) totheoppositeside(spot#58). Theresults showed different quantities of Sr and Mn between the two nacreous componentsexamined,withthespotsnumbered1–11and45–58withintheyellownacrerevealinghigherconcentrationsofSrthanMn,asexpectedforSWmaterial,whilespots12–44 showedhigherconcentrationsofMnthanSr(figure9), characteristicofFWmaterial.

Figure9. MacroimageofpearlAshowing58LA-ICP-MSspotstraversingtheSWyellow-orangenacreandtheinnerFWwhitenucleus. ThedifferencesinthequantitiesofMnandSrdetectedinthespotsacrossthetwoareasareshowninfigure11. PhotobySasithornEngniwat.

21

⁰

Figure10. MacroimageofpearlBshowingatotalof92LA-ICP-MSspotstraversingtheinnersurface. ThedifferencesinthequantitiesofMnandSrdetectedinthespotsacrossthetwoareasareshowninfigure12. PhotobySasithornEngniwat. TheLA-ICP-MSresultsonpearlBwereasclearasthoseobtainedfrompearlA. Thedifferentquantities of Sr and Mn within different areas of the saltwater nacre overgrowth andfreshwaternucleuswerereadilyapparent. Atotalof92laserablationspotswerepositionedintwolinesacrosstwoareasoftheinnersurface. Theresultsobtainedfromspots1–49and83–92provedthat theseareas formed inasaltwaterenvironment,whilespots50–82within thepearl’snucleus(creamcolorinfigure10) matchedwiththechemistryexpectedforafreshwaterenvironment. ThequantityofSrwashigherthanMninthenacreovergrowthareaandinthesolidnacreareaofthetwin(spots1–38) aroundthevoidtotheleftoftheboundaryline(redarrowinfigure 10),whilethequantityofMnwashigherthanSrinthefreshwaternucleus(figure10). Initiallytherewassomeconcernthattheareaaroundthevoid,thewhitering immediatelysurrounding it (seebluearrows),couldbeanotheraBCnucleus,but theLA-ICP-MS results confirmed that the chemistry was SW and hence more likely related to theformationofthelayersaroundtheFWnucleus,orwasaseparateNBC(“keshi”) formationthatmergedwith the SW layers surrounding the FW nucleus as the former aBC pearl developed (Krzemnickietal.,2011;Nilpetployetal.,2018).

22

⁰

PlottingtheLA-ICP-MSspots(figures11and12) clearlyshowsthattheMnlevelsoftheFWareas is considerablyhigher in the twonucleusareas,while thecorrespondingSr levelsdropsignificantly,aswouldbeexpectedinFWmaterials. TheoppositeistruewhentheSWareasareviewed.

Figure11. PearlAshowingthequantitiesofMnandSr(ppmw) inrelationtotheLA-ICP-MSspotstested(1–58) traversingthepearl(figure9). Spots12–44 fallwithinthefreshwaternucleusandcorrespondtohigherMnandlowerSrlevels.

0

1000

2000

3000

4000

5000

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58

SampleA

Sr

Mn

CONCE

NTR

ATION(P

PMW)

SPOTNUMBER

23

⁰

Figure12. PearlBshowingthequantitiesofMnandSr(ppmw) inrelationtotheLA-ICP-MSspotstested(1–92) traversingthepearl(figure10). Spots50–82fallwithinthefreshwaternucleuswhichcorrespondtohigherMnandlowerSrlevels.

4.ConclusionTheoverallappearanceofthe12SouthSeapearlsloanedforthisworkindicatedthattheywereprobablycultured. Theirexternalappearancewassimilar,withonlytheircolorsshowingsignificantdifference. Yettheoverallfeatureswereconsistentwithbeadculturedornon-beadculturedpearlsfromthePinctadamaximamollusk. Standardlaboratorypearlidentificationmethods,primarilyRTXanalysis,revealedthat11sampleswereaBCpearls,asclaimedbythesupplier,whileoneshowedastructurecharacteristicofNBCpearls. The11aBCpearlsshowedademarcationboundary typicalof suchpearlsbetween theouter surfacesof the freshwaterpearlnucleiandtheinnersurfacesoftheoverlyingsaltwaternacreovergrowth. However,thedefinition of the boundary between the two componentswas not always obvious, and thussomesamplesrequiredfurtherμ-CTanalysis. Itwasinterestingtonotethatthemoreelongatedbaroqueshapepearls,suchasthetwoselectedforfurtherdetailedanalyses(pearlsAandB), exhibitedmorepronounceddemarcationboundaries. Thiswaspartofthereasonwhybothsampleswereexamined furtherusingavarietyoftechniques. Microradiographyinitiallyrevealed internalstructuresindicatingaFWNBCpearl

0

400

800

1200

1600

1 11 21 31 41 51 61 71 81 91

SampleB

Sr

Mn

SPOTNUMBER

CONCE

NTR

ATION(P

PMW)

24

⁰

wasusedasanucleus. TheiropticalX-rayfluorescencereactionsalsoindicatedthatthenucleiwereFWorigin,sincethestrengthofthereactionwasgreaterthanthatnormallyobservedinpearlsfromaSWenvironment.DestructivetestingwassubsequentlyappliedtosamplesAandBinordertorevealtheir internalstructures moreclearly. Theactualboundariesinrelationtothe RTX results were then clearly visible, and the optical X-ray fluorescence reactions wereextremely telling and allowed the unquestionable separation of the two differentenvironmentalcomponents. Theirdefinedreactionsalsotalliedpreciselywiththetraceelementchemistrydatasubsequentlycollectedfromtheirinnersurfaces. Eventhoughstandardgemologicaltestingisusuallyenoughtoidentifysuchpearls, thesawingofthetwosamplesprovedtobeausefulexercisetocarryoutforthesakeofscientificexaminationandledtoevengreaterclarity.Thisstudyhelpedinclarifyingthecharacteristicstructuresexpectedinthistypeof atypical bead cultured pearl and in distinguishing it fromother entirely saltwater atypicalbeadculturedpearls(boththenucleusandovergrownlayersformedinSW) aswellasSWnon-beadculturedpearls. Thesawinghelpedfurtherourunderstandingofthepearls,andthesubsequentLA-ICP-MSanalysesprovedtobeapowerfultechniquethatallowedustotracethechemistryacrossthetwoformationenvironments. Whendestructivetestingisnotanoption,somepearl samplesmight exhibit questionable RTX resultswhere the demarcation betweenthetwocomponentsisunclear,andthusfurtherμ-CTexaminationmaybe necessary. Destructivetestingnaturallyhasits drawbacksandwasonlyappliedinthisresearchwiththevendor’sfullauthorizationsinceitultimatelychangesthepearls’shape,luster,andweight. Apearl’sinternalstructure isobviouslydamagedbytheprocessaswell. LA-ICP-MSspotsonlyaffectthesurfaceconditiontoasmalldegreeandresult inverysmallpartialdrillholesinthetestedareas. Theweightlossisinsignificant,andthusthemethodisacceptableinpearltestingwhennecessary. Virtuallyallpearltestingtechniquescarriedoutbygemologicallaboratoriesarenon-destructive. MethodssuchasRTX,opticalX-rayfluorescence,andμ-CTveryrarelyimpact theappearanceofanypearlsandare theonly techniquesneeded to identifyatypicalbeadculturedSouthSeapearlnucleatedwithfreshwaternon-beadculturedpearls.

5.AbouttheAuthorsMr. KessrapongandMs. LawanwongareanalyticstechniciansinthepearlidentificationdepartmentatGIAinBangkok.

6.AcknowledgmentsTheauthorsappreciatethecooperationofMr. DevchandChodhryofOrientPearl(Bangkok)Limited,wholoanedthesamplestoGIAandallowedgemologistsatGIA’sBangkoklaboratorytoconductdestructivetestingonthetwoselectedsamples. OthermembersofGIA’spearlidentification team, field gemology, and colored stone teams are also thanked for theirassistanceincollectingdataandproducingthisarticle.

25

⁰

ReferencesAbduriyim A. (2018) Cultured pearls from Lake Kasumigaura: Production and gemologicalcharacteristics. G&G,Vol. 54,No. 2,pp. 166–183,http://dx.doi.org/10.5741/GEMS.54.2.166Cartier L.E., Krzemnicki M.S. (2016) Golden South Sea cultured pearls: Cultivation steps &Gemological investigations. TheJournalofGemologicalAssociationofHongKong,Vol.37,pp. 16–21.

ElenS. (2001) Spectralreflectanceandfluorescencecharacteristicsofnatural-colorandheat-treated “golden” South Sea cultured pearls. G&G, Vol. 37, No. 2, pp. 114–123, http://dx.doi.org/10.5741/GEMS.37.2.114GemologicalInstituteofAmerica (2011) Pearls. TheBerneConvention,UnitedStatesofAmerica. HänniH.A.,KiefertL.,GieseP. (2005) X-rayluminescence,avaluabletestinpearlidentification. JournalofGemmology,Vol. 29,No. 5/6,pp. 325–329.HänniH.A.,KrzemnickiM.S.,CartierL.E. (2010) Appearanceofnewbeadmaterialinculturedpearls. JournalofGemmology,Vol. 32,No. 1-4,pp. 31–37. HomkrajaeA.,SunZ.,BlodgettT.,ZhouC. (2019) ProvenancediscriminationoffreshwaterpearlbyLA-ICP-MSandlineardiscriminationanalysis (LDA). G&G,Vol. 55,No. 1,pp. 47–60, http://dx.doi.org/10.5741/GEMS.55.1.47HsuT.,ZhouC.,HomkrajaeA.,HoJ.W.Y.,YazawaE.,PaduaP. (2016) FreshwaterpearlinginTennessee. GIAResearchNews,https://www.gia.edu/gia-news-research/freshwater-pearling-tennesseeKarampelasS.,MohamedF.,AbdullaH.,AlmahmoodF.,FlamarziL.,SangsawongS.,Al-AlawiA. (2019) Chemicalcharacteristicsoffreshwaterandsaltwaternaturalandculturedpearlsfromdifferentbivalves. Minerals,Vol. 9,No. 357,pp. 1–20. KessrapongP.,LawanwongK.,SturmanN. (2017) Pinctadamaculata (Pipi) bead-culturedblisterpearlsattachedtotheirshells. GIAResearchNews,April25, https://www.gia.edu/gia-news-research/pinctada-maculata-bead-cultured-blister-pearls-shells Kessrapong P., Lawanwong K. (2018) Atypical “bead”-cultured Pinctada maxima pearlsnucleatedwithfreshwaternon-bead-culturedpearls. Proceedingsofthe 2018GIAInternationalGemologicalSymposium,G&G,Vol. 54,No. 3,pp. 294–295.

26

⁰

KrzemnickiS.M.,Müller.,HänniH.A.,GutH.P.,DüggelinM. (2011) Tokkipearls: Additionalculturedpearlsformedduringpearlcultivation: Externalandinternalstructures. 32ndInternationalGemmologicalConference2011,pp. 1–14. KrzemnickiS.M.,FriessD.S.,ChalusP.,HänniH.A.,KarampelasS. (2010) X-raycomputedmicrotomography: Distinguishingnaturalpearls frombeadandnon-beadedculturedpearls. G&G,Vol. 46,No. 2,pp. 128–134,http://dx.doi.org/10.5741/GEMS.46.2.128KwakK.,LeeJ.,JeongE. (2016) IdentificationofdyedgoldenSouthSeapearlsusingUV-VisandPLtests. TheJournaloftheGemologicalAssociationofHongKong,Vol. 37,pp. 58–61. NilpetployN.,LawanwongK.,KessrapongP. (2018) Non-beadculturedpearlsfromPinctadamargaritifera. GIAResearchNews,April27,https://www.gia.edu/gia-news-research/non-bead-cultured-pearls-from-pinctada-margaritiferaScarrattK.,SturmanN.,TawfeeqA.,BracherP.,BracherM.,HomkrajaeA.,ManustrongA.,Somsa-ardN.,ZhouC. (2017) Atypical“beading”intheproductionofculturedpearlsfromAustralianPinctadamaxima. GIAResearchNews,https://www.gia.edu/gia-news-research/atypical-beading-production-cultured-pearls-australian-pinctada-maximaStrackE. (2006) Pearls. Rühle-Diebener-Verlag,Baden-Baden,Germany. SturmanN. (2009) Themicroradiographicstructuresofnon-beadculturedpearls. GIAResearchNews,https://www.gia.edu/gia-news-research-NR112009SturmanN.,HomkrajaeA.,ManustrongA.,Somsa-ardN. (2015) X-raycomputedmicrotomography(μ-CT) structuresofknownnaturalandnon-beadculturedPinctadamaximapearls. Proceedingsofthe34thInternationalGemmologicalConference,Vilnius,Lithuania.SturmanN.,BergmanJ.,PoliJ.,HomkrajaeA.,ManustrongA.,Somsa-ard.N. (2016) Bead-culturedandnon-bead-culturedpearlsfromLombok,Indonesia. G&G,Vol. 52,No. 3,pp. 288–297, http://dx.doi.org/10.5741/GEMS.52.3.288SturmanN.,OtterL.M.,HomkrajaeA.,ManustrongA.,NilpetployN.,LawanwongK.,KessrapongP.,JochumK.P.,StollB.,GötzH.,JacobD.E. (2019)Apearlidentificationchallenge. G&G,Vol. 55,No.2,pp. 229–243,http://dx.doi.org/10.5741/GEMS.55.2.229UrmosJ.,SharmaS.K.,MachenzeF.T. (1991) CharacterizationofsomebiogeniccarbonateswithRamanspectroscopy. AmericanMineralogist,Vol. 76,pp. 641–646.

27

⁰

ZhouC.,HomkrajaeA.,HoJ.W.Y.,HyattA.,SturmanN. (2012) Updateontheidentificationofdyetreatmentinyellowor“golden”culturedpearls. G&G,Vol. 48,No. 4,pp. 284–291,http://dx.doi.org/10.5741/GEMS.48.4.284