DistinguishingHeatedfromUnheatedNaturalSpinels

Ashortreviewofongoingresearch

SudaratSaeseaw,WuyiWangandKennethScarratt
GIALaboratories,BangkokandNewYork(March22,2009)

Figure1:SpinelroughfromTajikistan,thelargeststone(upperright)weighing242.50ct.

Abstract

Naturalspinelsarefoundincolorsequaltothoseoffinerubiesorsapphires,andsince
topqualityspinelsareoneofthefewbetterknowngemstohavesofarescapedthe
attention of large scale treatment processes; they are gaining a greater commercial
importance. However, it would appear that at least some heat treated natural spinels
are on the market and therefore it is important that criteria be established for the
separationofthetreatedfromthenaturalmaterial.Thispaperreviewsandconfirmsthe
identification criteria developed during experiments carried out in 2005 by John L
Emmett(ofJLEAssociateslocatedinBrushPrairie,WA)andWuyiWangonpinktored

spinel from Tanzania and the application of these to spinels from other sources
examinedbytheGIAsincetheoriginalexperiments.

Figure2:FacetedTajikistanspinelscutfromtheroughdepictedinFigure1.

Introduction

In2005someconcernwasexpressedbythetradethatcertainpinkandredspinelsfrom
Tanzania may have been heated to improve their appearance. In order to investigate
theseconcernsfurtherfivesampleswereobtainedfromthetradethatwerestatedto
be unheated. These five were examined and their data recorded and then subject to
heatingprocessesattemperaturesvaryingfrom1100to1700C,inoxidizingconditions,
byJohnEmmett.

Ingeneral,andforthefivesamplescited,theseexperimentsshowedthatthespinelsdid
not improve in appearance but rather they became darker (Figure 3). However, the
experiment was limited to a very small sampling. Before and after inclusion
observations revealed some severe alterations had taken place following heating,
particularlyatthehighertemperatures(Figure5andFigure6).

Figure3:A1.17ctspinelfromTanzaniabeforeheating(left)andafterheatingat1400C,inO2,for5hrs(right).

Figure4:Theinclusionsceneina1.17ctspinelfromTanzaniapriortoheating(seealsoFigure3andFigure5).


Figure5:Theinclusionsceneina1.17ctspinelfromTanzaniaafterheatingat1400C(seealsoFigure3andFigure4)

Whiletheseheatalteredinclusionscenesundoubtedlyhelpinidentifyingthefactthat
heating has taken place, further criteria were needed for stones that did not contain
inclusions natural spinel is often free of inclusions. The investigation rapidly
progressed to observing photoluminescence (PL) spectra where it was felt key criteria
may exist. Here it was discovered that distinct and typical changes occur in the PL
spectraofpinktoredTanzanianspinelfollowingheating.

Usinga488laser(the514laserwillproducesimilarresults)aseriesofrelativelysharp
peaksarerecordedinPLbetween600and700nminunheatedTanzanianspinel.Inall
cases, following heating these sharp peaks become much broader and less defined
(Figure 7). Thus simple criteria emerged for distinguishing heated from unheated
Tanzanianpinktoredspinel.

Figure6:AnexampleofheatalteredinclusionsinaTanzanianspinelfollowing5hoursat1250C.

Figure7:ThePLspectrumofaTanzanianspinelbeforeandafterheatingat1100C.Notethedistinctbroadeningofthepeaks
followingtreatment.


Figure8:AselectionoftheBurmeseoriginspinelsfromTheGbelinCollectionexaminedsincethe2005investigationtoconfirmthe
originalPLidentificationcriteriaforheatedspinel.

Furtherexaminationofthecriteria

ThePLspectrafor108naturalunheatedBurmesespinelsfromtheGbelinCollection 1
(alongwith13syntheticspinelsbothflamefusionandflux/pulledcrystals),aswellas
bothroughandcutpinktoredspinelfromTajikistanwereexamined.Forthenatural
pink to red spinels the typical sharp peak PL spectra were recorded for all stones.
ExamplesaredepictedinFigure9,Figure10,andFigure11.

PL spectroscopy therefore seems to be an invaluable tool in separating heated from

unheatedpinktoredspinel.However,itshouldalsobenotedthatsimilarbroadrather
thansharpPLspectraareobservedinsyntheticpinktoredspinel.Thatis,differences
observed between fluxgrown synthetics and natural pink to red spinels, also center
around the sharpness of the PL peaks. In Verneuil synthetic spinels this broadening
characteristic is reportedly due to structural disorders resulting from excess Al, and it
has also been reported that the broader bands in other types of synthetic spinel (i.e.

AlargecollectionofgemstonespurchasedfromtheestateofDrEdwardGbelinbyGIA.

fluxgrown)arealsoduetothepresenceofirregularities(disorder)anddefectsintheir
crystallattice(ShenA.H.,2004,Krzemnicki,2008).

Figure9:ThistypicalPLspectrumwasobservedinmostofthenaturalspinelsfromBurmatested,withastrongandsharpemission
peakat685.6nm.Among108samplestested,only4samplesdidnotshowthisCrrelatedemissionline.These4samplesareall
black(2fromSriLankaand2fromThailand).ConcentrationsofCrinthesestonesareatasimilarleveltothatdetectedinmost
othernaturalspinelstested,buttheFeAl2O4componentisveryhigh(3537mol%).

Pinktoredspinelsthathavebroadratherthansharppeaks(Figure7)intheirPLspectra
maythereforebeassumedtobeeitherofsyntheticoriginorheatednaturalstones.In
such cases inclusions and or chemistry may be needed to conclude identification
(Muhlmeister,etal.,1993).

EnergyDispersiveXRayFluorescence(EDXRF)traceelementanalysisisanexcellenttool
forseparatingsyntheticfromnaturalspinels.Influxgrownsyntheticspinels,chromium
(Cr)andiron(Fe)arefoundtobemajortraceelementsandnickel(Ni),vanadium(V),
zinc (Zn), and gallium (Ga) are all found to be minor trace elements. In contrast, all
natural spinels analyzed thus far, have had higher levels of Zn than the fluxgrown
synthetics,andTiispresentasaminortraceelement.

LAICPMSanalysisrevealedsignificantamountsofLi,Be,Zn,andGainallofthenatural
spinels(over100samples)tested.Theseelementsareusuallyabsentorextremelylow
in concentration in synthetic spinels. No overlap in concentration was observed
betweennaturalandsyntheticspinels(Figure12andFigure13).


Figure10:TypicalPLspectrumfoundforsamplesofalargeparcelofpinktoredspinelfromTajikistan(seeFigure1).

Figure11:TypicalPLspectrumfoundforsevenpinktoredspinelfromTajikistan(seeFigure2).

Figure12:Traceelementpopulationfields(GaandZn)fornaturalandsyntheticspinelshowingaclearseparationbetweenthetwo.

Figure13:Traceelementpopulationfields(BeandLi)fornaturalandsyntheticspinelshowingaclearseparationbetweenthetwo.

Conclusion

EDXRF chemical analysis and photoluminescence (PL) spectroscopy are crucial in

separating clean examples of natural from fluxgrown synthetic spinel. Whilst some
inclusionfeatures(Figure4,Figure5,andFigure6),whenpresentanddependingonthe
typeandquantity,mightassistinseparatingunheatedfromheatedspinels,itwasfound
that PL spectroscopy (Figure 7) was also crucial in determining whether sampleswere
heatedorunheated.

Furtherandmorecomprehensivedatawillbepublishedasresearchcontinues.

References:

Koivula, J. I. and Kammerling, R. C. (1994) The characteristics of Russian flux-grown synthetic red and
blue spinels. South African Gemmologist. 8. 2/3. 4-15
Krzemnicki, M. S. (2008) Trade Alert: Flux Grown Synthetic Spinels Again on the Market. SSEF
Newsletter.
Muhlmeister, S., Koivula, J. I., Kammerling, R. C., Smith, C. P., Fritsch, E. and Shigley, J. E. (1993)
Flux-grown synthetic red and blue spinels from Russia. Gems & Gemology. 29. 2. 81-98
Shen A.H., B. C. M., DeGhionno D. (2004) Lab Notes: Natural spinel identified with Photoluminescence.
Gems and Gemology. 40. 2. 168-169.

Furtherreading
Bank, H., Lenzen, G. and Henn, U. (1991) Neue Edelsteinvorkommen - Neue Synthesen. Gemmologie
Aktuell. 3.
Brown, G., Kelly, S. M. B. and Sneyd, R. (1990) Russian flux-grown synthetic spinel. Australian
Gemmologist. 17. 8. 315-317
Crowningshield, G. R. and Holmes, R. J. (1950) Synthetic red spinel. Gems & Gemology. 6. 12. 362-368
Crowningshield, R. (1971) Red synthetic spinel. Gems & Gemology. 13. 11. 350-351
Eppler, W. F. (1953) Further observations on synthetic red spinel. Gems & Gemology. 7. 10. 306
Gbelin, E. J. (195253) More news of synthetic red spinel. Gems & Gemology. 7. 8. 236247
Gbelin, E. J. (1954) Synthetischer roter Spinell. Zeitschrift der Deutschen Gesellschaft fr
Edelsteinkunde. 4
Hodkinson, A. (1991) Synthetic red spinel. Australian Gemmologist. 17. 11. 466-468
Johnson, M. L. and Koivula, J. I. (1997) New information on flux-grown red spinel from Russia. Gems &
Gemology. 33. 2. 151-152
Kitawaki, Y. (1994) Synthetic red spinel. Journal of the Gemmological Society of Japan. 19. 1-4. 34
Koivula J.I, K. R. C. and Fritsch, E. (1993) Synthetic spinel from eastern Germany. Gems & Gemology.
29. 2. 140-141
Koivula, J. I. and Kammerling, R. C. (1989) Flux synthetic spinel. Gems & Gemology. 25. 4. 250