American M ineralogist, Volume60, pages 705-709,1975

Blue and Brown Topaz Producedby Gammalrradiation Kunr NlssAU ANDBrrry E. pnnscorr Bell Laboratories,Murray Hill, New Jersey07924 Abstract The well-knownbrown color producedby the irradiationof topazhasbeenreexamined, in particularwith respectto the kineticsof the color formation.The known color is producedat two differentrates:fast colorationwith a half colorationdoseof 0.058megaradsof 7 rays from cobalt60 (15 min in the cell)producesa yellowto "sherry" color which fadesrapidly on exposureto light; the slowcolorationwith a half doseof 12.6megarads(18hours)saturatesto a dark brown which fadesmore slowly.Not all specimens go throughboth stages.Heatingto 200'C for a few hours removesessentiallyall the color of the specimens tested.Topaz from Mexico and somefrom Brazil appearto be most resistantto light fading. Somerarecolorlesstopazspecimens (of unknownorigin but probablyfrom Brazil)afterirradiationand heatbleachingbecomeblue.This colorationis stableto light and heatand indistinguishable by standardgemologicaltestsfrom natural blue topaz,both showingan absorption band at 0.62 pm.

Introduction

(Dickinson and Moore, 1967; Samoilovichand Novozhilov,1965).Chromiummay be the colorant The observationthat topaz turns yellow, ran, in thoseyellow topazesthat turn pink (possiblythe orange,or brown on irradiationby radium emana- changeCr6+* Cr3+?)but certainlynot in most other tions,X rays,gammatays,etc, as well as the rapid yellow topazes which usually lack significant fading of this color on exposureto light or heat, has amountsof chromium.Iron hasbeensuggested asthe been frequently noted (e.g., Wild and Liesegang, causeof the blue color in topaz(Wild and Liesegang, 1923;Lind and Bardwell, 1923;Pough and Rogers, 1923;Howard, 1935;Lemmleinand Melankholin, 1947; LieIz and Klug, 1956).That some of this 195l), as has a color center(Dickinsonand Moore, materialalsodevelopsthe 0.62pm absorptionband 1967). characteristicof natural blue topaz on irradiation The electron paramagneticresonancework. of and becomesblue whenthe accompanying Dickinson brown is and Moore (1967) is most significant. removed by heating does not, however,appear to They recognizedIen distinct EpRcenters,four from have been previouslynoted. impurities(two Fe3+,Va+,and Cr3+),and six from The understandingof the causesof the various defectcenters(designated A, B, X, Y, Z, and F). The colorsin topazis not well-developed. The only unam- Cr8+centerwasidentifiedwith pink, but Fe8+centers biguous assignmentis that of chromium, which occurredin colorless,blue, and yellow topaz, and causespink and red,aswell asthe reddishcomponent henceare probablynot color causing.Unfortunately in someorangeor "sherry" coloredmaterial(someof Fe2*is not seenin Epn.The A and B defectcenters, which turns pink on heating).Chromium as the both of the trapped electrontype, appearedto corcolorant was first proposedby Wild and Liesegang relatewith the natural yellowcolor lost by heatingto (1923) and later confirmed by optical (Grum- 500"Cbut recoveredon irradiation.Y is a holecenter Grzhimailo, 1953) and paramagneticresonance associablewith a yellow color producedby further (Dickinsonand Moore, 1967)spectra.Vanadiumhas heatingof thisnow colorless materialto 1000"C.The also beensuggested as the causeof pink (Howard, hole center designatedX correlatedwith the blue 1935),as has Fee+ (Lemmlein and Melankholin, color (lost by heatingto 500"C),and the hole center r9sI ). designatedF was correlatedwith the intensityof the Natural yellow and brown colors in topaz have irradiation-inducedyellow-orangeobserved in been attributedto iron (Grum-Grzhimailo,1953), crystalsof any.color and lost again on heatingto chromium (Howard, 1935), or to color centers 500'C.This last@nter,havinga g valueof about 2.09, 't05

K. NASSAU, AND B, E. PRESCOTT

706

wasinterpretedas possiblybeinga hole trapped"on that oxygenof a silicontetrahedronwhich is coplanar by six fluorideions hexagonally with and surrounded in the (010)crystalplane" (Dickinsonand Moore, t967). (1965,1969)alsoused and co-workers Samoilovich Epn to study irradiation induced color centersin topaz.They correlateda resonanceat g : I '982with the depthof the color and the amountof irradiation (up to a saturationlimit), and also interpretedthe defectas an F centertype, consistinghoweverof an electrontrappedat a fluoridevacancy.This is clearly inconsistentwith the F centerinvolving a hole describedby Dickinsonand Moore (1967)'

FIc.

l. Cleavage fragments of originally colorless topaz showing unevennessof coloration upon irradiation.

The lrradiation InducedBrown Color in Topaz Experimental acquireda more-or-less As expected,all specimens (Tablel), the coloration to brown yellow to orange were examined, A total of I23 topaz specimens color, the the original on depending Exshade exact somein crystalform and someascut gemstones' suscepgamma rays, and the to of exposure length cept for five crystals from Mexico and two from varied last This coloration. to material of the Topaz Mt., ThomasRange,Utah, thesespecimens tibility shown are Figure I In a crystal. within greatly even unwere either from Minas Gerais, Btazil, or of known origin, probably also from BraziI.The initial two topaz cleavagesectionsoriginally colorless,ircolors usedand the resultsobtainedare summarized radiatedwith 2 megarads.Both striationsand patches of brown have developed,some of which follow in Table l. gamma crystalfacesand someof which do not. in a Co-60 Irradiations were carried out per The polarizationand the rateof colorationwith ir0.70 megarads of cell with an irradiationintensity dosewere studiedon a near cube approxradiation specieach on hour. At least l0 megaradswas used 8 mm to a side, oriented with the crystalimately furnace in a muffie out men. Heating was carried The threepolarizedspectraareshown axes. and lographic heated and in sand buried with the samples for two differentlevelsof irradiation(15 2 in Figure prevent cracking. to cooledslowly : Polarizedoptical spectrawererun on a Cary l4R min : 0.175megarads,20 hours 14 megarads). facetedstoneswereimmersedin The basicchangeis the shift of the absorptionedge, spectrophotometer; near 0.3 u.m before irradiation, acrossthe visible in a fusedsilica cell. tetrachloride carbon region,to producethe brown coloration.The only structureseenis the absorptionband on the side of E parallelto TnsI-e l. Color Changeson the Gamma lrradiation of 123 the edgenear0.35pm in the spectrawith (Fig. 2) pleochroism with l0 Megarads the c axis.There is pronounced Topaz Specimens p )) (med. brown). 7 (tan). with a (dark brown) > No. ot Ini ti al pm as a at'0'5 Using the absorptioncoefficient Color Irradiated Specimens Color irradiawith of coloration measureof the intensity 28 tion dose,the data points of Figure 3 wereobtained. No change to pale-yellow Colorl ess zo broun Med. yeLlow to light This is for the polarizationwith E parallel to the D 32* Brom to dark brown axis, but the other orientationsgave essentiallythe t4 No change to darker yellow Pale Yellow sameresults.As can be seenthere is an initial rapid t7 Mediun to dark sherry coloration,saturatingin lessthan onehour, aswell as 1 darker Slightly Sherry a muchslowercomponent,saturatingonly after some (with days (the "infinity" point was measuredafter a 30 or an olive Pale brown BIue to Green 2 day irradiation). Mediun to dark brom An exponentialtype coloration equationis: craani

ch

+i nf

I

*HeatLng of these pz'oduced a blue coLot' i-n 21 specimens in aLL tVe some of uhich tnd qn olitse oz' greeni'sh tint; otVer 102 specimens heating nerely restoned the on'Lgi-nal . coLon,

dmax -

ott = o,maxexP (-

trrt)

(l)

BLUE AND BROWN TOPAZ BY GAMMA IRRADIATION

dmax - @d : dmax exp (-

where amaxis the saturation absorption coefficient,a1 or aa is the same coemcient after irradiation time t or dose d, and \s and tra are constants. Rewriting this in terms of the half coloration time. lr7, where: t112= 0.693/\1

ot"^r, -d"^u*

d'^

a"

02 0l

F

DOSE IN MEGARAOS

o

E

30

T|MErN MINUTES

20

d F

z

dvz = 0.693/)tt (4) ' ', and using superscripts and for two separare colorationprocesses occurringsimultaneouslygives: ol'^ax I

--*.ro*,

o7

(3)

or the half coloration dose,dr,r, where

dt :

olu-1

(2)

trdd)

70'l

* exp (-0.693

z

to

t/tr,")

exp (-0.693 t/t'rrr)

/--2.'d'l

(5)

FAsr|

and d.d = q.tmax+ ot"^ *'_e'^^* -ot"n",

exp (-0.693

exp (-0.693

d/d,r/2)

dld,r,r)

(6)

Frc. 3. Developmentof the color in the initially colorless topaz of Figure2 on gammairradiation.points areexperimental; the curvesare calculatedfrom Equation(5). The..infinity" point was the result of irradiation for 720 hours.

By plotting on a logarithmicscale(Fig. a) the two componentswere separatedand gavefor the data of Figure 3 the valuesshown in Table 2 (more precise Table 2 and can be seento representthe two colorathan thosegiven by Nassau,l9Z4).The solid curves tion processes well within the experimentalerror. in Figure 3 were drawn using theseparametersof Color photographscorrespondingto the two stagesof colorationof Figures2 and3 werepublished by Nassau(1974).The color after 15 min is a warm "sherry" brown; after 20 hours, the color is much darker "cinnamon" brown and lessattractive.These two colorationstagesalsoreactdifferentlyto light exposure;the rapidly formed sherrybrown color (yellow if only a very short irradiation)fadesafter a I or 2 day exposureto sun/daylight, while the slower forming but darker color requiresseveralweeksto

1

td o

z

T I M E ( H o U R s) 15 20

@ E

o

o (D :lo

fioe O"= O.c. - cr

[06 =o4

tljz-roxouas

=gA

2

0.3 Ftc.

0.5

0.7

0.5 0.5 0.7 0.3 WAVELENGTH tN /l,m

0.5

0.7

2. Polarized absorption curves of a colorless topaz that turned brown on gamma irradiation.

0.2

4 - =d r - a ' t l;,2

oro20

5 MINS.

T I M E( M I N U T E S )

Flc. 4. Logarithmic decompositionof the data of Figure 3 into two comDonents.

K. NASSAU. AND B. E, PRESCOTT

708

for the Topaz of Figures2 and 3 TlsLe 2. ColorationParameters Rapid Coloration saturation coefficient

A b s o r p-t i o n

= q' max

0.40

= 5.0 nins.

coloration

Time

til2

Half Coloration (Megarads)

Dose

dtr/2 = 0.058

HaIf

slow co lorat ion d't max

=

tir,

= 18 hours

1.60

d'i/2 = 12.6

fade. Most topaz was found to go through both stages.However someyellow specimensfrom Brazil (e.g.,the @nter stone,Fig. 9, Nassau,1974)only showed the rapid coloration even on extendedirradiation and faded very quickly. Some colorless topaz from Brazil, as well as colorlesstopaz from Mexico(e.g.,centercrystal,lowerrow, Fig. 10,Nassau, 1974),turned very dark and still retainedsignificant color after some months of exposureto light. Heatingto as little as200'C for a few hours removed all this irradiation-inducedcolor. Somevariation not only in the saturationcoloration but alsoin the half colorationLimetuzwasnoted

1

U z @

@

o.2

0.3

0.6 0.4 0.5 lN Pm WAVELENGTH

0.7

0.8

Frc.5. Unpolarizedabsorptioncurvesof (a) a colorlesstopaz turned brown after 20 hours irradiation; (b) same as (a) but after it turned blue through heatingovernight at 200'C; (c) a natural blue topaz.

in the caseof the Maxixe type deep-bluecolor center in beryl (Nassau,19?4),and somevariationmay also be anticipatedin topaz. Thus the saturationcoloration of - 1.5X t0?R of Samoilovichand Chentsova (1969) can be consideredas being in reasonable agreementwith our drn of 12.6megarads. Samoilovichet al (1965,1969)obtaineda brown color with a saturationdoesof 15 megaradsin their electrontype F center topaz, while Dickinson and colorobMoore (1967)noteda "pale amber-yellow" tainedwith 5 megaradsin their initially colorlesshole would not be type F centerlopaz.Theseobservations inconsistentwith each other and with our current results, if the hole type F center with g ) 2 of Dickinson and Moore correlateswith the rapidly producedyellow to sherrycolor, and if the electron type F centerwith g 12 of Samoilovichet al (1965, 1969)correlateswith the more slowlyproduceddark brown color. This is merely a suggestionfor the and clearlyneedscorresolutionofthis inconsistency roboratingdata. Crowningshield(197a)noted that one of theseirradiateddark brown topazcrystals(origin unknown) when partiallylight bleachedbecame"smoky," !.e., similar in appearanceto smoky quartz without a significant brown component, an apparently previouslyunrePortedcolor. The lrradiation InducedBlue Color in Topaz This was first noted during a study of the development of the smoky color and its destructionby heating in a group of colorlessquartzstonesloanedby R. Crowningshieldof the GemologicalInstitute of America,when one stoneturnedan olive or greenishbrown on irradiation and becameblue on heating overnightat 250oC.Testsshowedit to be topazand not quartz. The spectrumof this stone,which had beeninitially colorless,is shown in Figure 5 both before and after reirradiation and compared with a natural topaz.The blue color in both is due to a broad abBoth by this testand by standard sorption aLO.62pm. gemologicaltests (e.g., Webster, 1972; Liddicoat, 1972) carried out by R. Crowningshield,this blue irradiation-producedcolor proved to be indistinguishablefrom naturally-occurringblue topaz' It also behavedon exposureto light in the sameway as much natural blue topaz,fadingslightlyin the first two daysbut showingno significantfurther changein the next two months. Heating to 250'C overnighthad no effecton the blue color. but at 500"C decolorationoccurredas it

BLAE AND BROWN TOPAZ BY GAMMA IRRADIATION

709

doesin natural blue topaz (Dickinsonand Moore. tendencyto cleaveduring heating after irradiation, 1967;Lemmleinand Melankholin,l95l). A natural and (b) that topaz which is alreadypale blue turns deep blue topaz, which had paled significantlyon darker blue only very rarely. possible A deduction light exposure,did not returnto the originalcolor but from the latter observation is that natural blue remainedpale after the irradiation treatment.When topaz may have becomeblue from natural irradiacompletelybleachedto colorlessby heatingto 500oC, tion; sincethe accompanying brown is not usually ,theblue color could be partiallyrestoredby gamma seen,this also impliesthat the accompanyingbrown irradiation followed by gentle heat treatment at color bleachedat ambient temperatureover geo200"Cfor a few hours.Color photosofthesechanges logicaltime periods. have beenpublishedby Nassau(1974). Acknowledgments Of the 86 colorlesstopazspecimens in Table l, 2l We wish to thank D. L. Wood for helpful advice on the wereturnedblue by this treatment.The effectis much aspectsof this study,and R. L. Crowningshield for rarer than this would indicatesincethe 2l blueswere spectroscopic usefuldiscussions and the gemologicaltestingof the blue topaz. found in the first 48 colorless specimenstested We are gratefulto the following for the loan of specimensfor (mostly from old collections),and no more were testing:Mr. R. Crowningshieldof the Gemologicallnstitute of found in the next 38 colorlessspecimens tested(from America,Mr. A. Heinzmannof H. R. Benedictand Sons,and Mr. more recentsources).It is likely that only material M. Strump of the SuperiorGem Company,all of New york City; from one or a few localitieshasthe specificprecursor Mr. E. C. Parrott of Vancouver,Washington;Mr. F. Boeseof Burlington,North Carolina;Mr. J. Baskins to this blue color, which is in all probabilitythe hole Middlesex,New Jersey;Mr. A. R. plart of Baskinsand Sons, of The Craftsmen,West typeX centerof DickinsonandMoore (1967).Anex_ End, New Jersey;and Mr. R. C. Romanellaof Commercial planationfor the lack of success in finding additional MineralsCorp., Scottsdale,Arizona. colorless material which would turn blue may References perhapsbe found in the report by Crowningshield R. (1974)Commercialimplicationsof gamma (1974)that gem grade blue topaz has recentlybeen CnowNtNcsHrer-o, radiationon gem materials.Z. Dt. Gemmol.Ges.23,95-101. appearingin abundanceand that some of this is DrcrrNsoN, A. C., lr.ro W. J. MooRE (1967) paramagnetic statedto havebeentreated:presumablyby the gam_ resonanceof metal ions and defect centersin topaz. J. Phys. Chem.7l, 231-240. ma ray process here described.Apparently this processhas been discoveredmore than once. Cin- Gnuu-Gnzsrl,rerlo,S. V. ( 1953)Nature of color of roseand yel(Mbm. Soc.Russe namon brown topaz also seemsto be becomine low topaz.Zapiski Vses.Mineral. Obshchestu. Mineral\, t2, 142-146. widely available. Hownno, J. W. (1935)Topaz.J. Chem.Educ. 12, 133-156. Summary Two distinct types of yellow to orangeto brown color in topaz,producedby gammairradiation,have beenobservedwith differentformation rates.One or both may occurin a specimen,and they may relateto two different hole and electrontype F centerspreviously observedin topaz by Ern. There is considerablevariation in the rate of fading on exposure to light. Somecolorlesstopaz becomesblue after irradiation followed by gentleheating.This transformation has not been previouslyreported,and the resulting material appearsto be indistinguishablefrom naturally occurringblue topaz.

LruuLrrN, G. G., ANDN. M. MeleNrsoux (1951)The coloration of dichroic topaz crystals. Trudy Inst. Kristallogr. Akad. Nauk

sssn, 6, 245-254.

Ltootcont, R. T. (1972) Handbook of Gem ldentifcation. GemologicalInstituteof America,Los Angeles,9th ed. Lrcrz, L, rNp H. G. Kr-uc (1956) Coloring and thermoluminescence of irradiated topaz.NeuesJahrb.Mineral. Abh. 9,27-40. LrNo, S. C., lNo D. C. BlnowBLL (1923) The coloring and producedin transparentmineralsand thermophosphorescence gemsby radium radiation.Am. Mineral. t, l7l-181. NAssAU,K. (1974)The effectsof gammarayson thecolor of beryl, smokyquartz,amethyst,and topaz.LapidaryJ.28,20-40; Correction,2E, 556. Poucs, F. H., er.roT. H. Rocrns (1947)Experiments in X-ray irradiationof gem stones.Am. Mineral. 12, 31-43. Slrvrorlovrcu,M. I., eNp L. G. CnrNrsovl (1969)Nature of irradiated topaz color. Tr. Vses.Nauch.-Issled.Inst. Sin.Mineral. Syr'ya, 1969, 129-130. -, Note Addedin proof AND A. L Novozurlov (1965) Electron paramagnetic The testing of over 100 additional specimens resonancein irradiatedtopaz.Zh. Strukt. Khim. 6, 461-463. Wrnsren, R. (1972) (pebblesfrom Brazil, courtesyof R. V. Gainesof Wrlo, G. O., .,rNo Gerns. Butterworths, London, 2nd ed. R. E. LrrseclNc (1923)The color of topaz. Pottstown,Pennsylvania) has confirmedtwo factors Centr. Mtneral. Geol. 1923. 385-387.

which had previously been suspected:(a) that topaz which turns blue showsan unusuallystrong

Manuscript receiued,Nouember14, 1974; accepted for publication March 21, 1975.