Luminescence behavior of turquoise [CuAl6(PO4)4(OH)8·4H2O]

Radiation Measurements 45 (2010) 749–752

Contents lists available at ScienceDirect

Radiation Measurements journal homepage: www.elsevier.com/locate/radmeas

Luminescence behavior of turquoise [CuAl6(PO4)4(OH)8$4H2O] E. Crespo-Feo a, *, J. Garcia-Guinea a, V. Correcher b, P. Prado-Herrero b a b

Museo Nacional de Ciencias Naturales, CSIC, C/Jose´ Gutie´rrez Abascal 2, 28006 Madrid, Spain CIEMAT, Av. Complutense 22, 28040 Madrid, Spain

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 August 2009 Received in revised form 16 November 2009 Accepted 20 December 2009

We, herein, study the thermoluminescence (TL) and cathodoluminescence (CL-SEM) emissions of a commercial turquoise to determine its possible use as an emergency dosimeter. CL spectrum of bulk sample displays an intense broad emission from w260 to w650 nm together with a weaker narrow band at w710 nm. Through EDS and EMPA chemical analyses, an important amount of rare earth elements (REE) such as Ce, La, Y, Nd, Dy, Yb, Er, Pr, Sm, Gd, Ho, Tb, and Tm have been identified associated with phosphate phases as well as in turquoise itself. Apatite [Ca5[OH(PO4)3]], monazite [(Ce,La,Nd,Th)PO4], and xenotime [YPO4] have been detected in the turquoise matrix, as well as very small amounts of quartz [SiO2]. Turquoise has a high content in Zn2þ substituting for Cu2þ, together with small amounts of Cr3þ substituting for Al3þ. The well defined peak observed at w710 nm can be due to Cr3þ activation centers in the turquoise lattice with the rest of the emissions associated with a REE activators and intrinsic defects. Regarding the blue TL, two emission bands (190 and 330  C) appear in the natural aliquot whereas the irradiated sample displays no response at all. This glow emission could be due to the losses of structural water molecules in the turquoise lattice. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Turquoise CuAl6(PO4)4(OH)8$4H2O Thermoluminescence Cathodoluminescence Emergency dosimeter

1. Introduction Turquoise is one of the oldest gemstones known in history, owing to its unique color, easy lapidary, and wide distribution. Ancient cultures in Egypt, Persia as well as Aztecs and Incas mined, traded and venerated this mineral. This opaque, blue-to-green mineral is a hydrated copper aluminum hydroxy phosphate, with the chemical formula CuAl6(PO4)4(OH)8$4H2O. Its crystal structure is triclinic, with space group P1 and it was defined by Cid-Dresdner (1965) and more recently by Kolitsch and Giester (2000). Some iron is usually present in natural samples substituted for Cu2þ. Indeed, the color of the turquoise seems to be related to the Cu/Fe ratio, showing bluer tonality at higher Cu/Fe concentrations meanwhile lower ratios give greenish areas (Clark et al., 1979). In general, Al3þ sites are predisposed to be substituted for luminescent cations such as Cr3þ, Mn2þ, Mn4þ, and scarcer for Ti4þ. As far as our knowledge, turquoise has been well characterized by spectroscopy techniques (Hunt et al., 1972; Frost et al., 2006) and by ESR and EPR (Diaz et al., 1971; Clark et al., 1979; Sharma et al., 1988), but no previous works on turquoise luminescence has been performed. Only Belik and Çiftçi (2008) published a report on turquoise cathodoluminescence though no conclusive results can

* Corresponding author. E-mail address: [email protected] (E. Crespo-Feo). 1350-4487/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.radmeas.2009.12.027

be obtained from their work since they worked on odontolite, a turquoise mineral imitation (see Reiche et al., 2000). As a general rule, luminescence lines in natural gemstones are broader than in their artificial counterparts. The reason is that many optically active centers are usually present in the natural gemstones, while only one or two occur in the artificial ones (Gaft et al., 2005). Luminescence study of turquoise and its physicochemical characterization were performed by X-ray diffraction (XRD), environmental scanning electron microscopy (ESEM-EDS), differential thermal analysis (DTA), thermoluminescence (TL) and cathodoluminescence coupled to an ESEM (ESEM-CL) techniques. Through its luminescence behavior, the possible use as an emergency dosimeter was determined since gems exhibit a reasonable sensitivity to radiation based on their luminescence (TL) properties. 2. Experimental The turquoise mineral was characterized by X-ray powder diffraction using a Phillips PW1710/00 diffractometer with a CuKa radiation source, equipped with a graphite monochromator. Patterns were obtained by step scanning from 2 to 64 (2q in steps of 0.020 ; 4 s per step) and compared with the XRD card files of the Joint Committee on Powder Diffraction Standards. The morphology and composition of turquoise, and of the associated less common minerals, were assessed under environmental scanning electron microscopy and energy dispersive X-ray spectroscopy (EDS–ESEM)

750

E. Crespo-Feo et al. / Radiation Measurements 45 (2010) 749–752

using an Inspect-S ESEM of the FEI Company. CL spectra were performed on a polished slab in low vacuum mode without coating, using a Gatan MonoCL3 detector and PA-3 photomultiplier attached to the ESEM. The PMT covers a spectral range of 185 nm– 850 nm, and is most sensitive in the blue parts of the spectrum. A retractable parabolic diamond mirror and a photomultiplier tube are used to collect and amplify luminescence. The sample was positioned 16.2 mm beneath the bottom of the CL mirror assembly. The excitation for CL measurements was provided at 25 kV electron beam. Electron microprobe analyses (EMPA) were performed on a polished slab in a Jeol Superprobe JXA-8900 M, with bulk and channel-selected (TAP, PETJ, LIF, PETH) X-ray spectra. Natural standards and synthetic have been used for this purpose. DTA–TG of 10 mg of crushed turquoise was recorded with a simultaneous DTA–TG thermal analyzer (Perkin Elmer, STA 6000) in nitrogen atmosphere at a heating rate of 10  C min1 from room temperature up to 1000  C. The sample was held in an alumina crucible and the reference material was platinum. Natural and irradiated glow curves were obtained using an automated Risø TL system model DA-12 provided with an EMI 9635 QA photomultiplier and a blue filter peaked at 320–480 nm, with 80  16 nm FWHM and 60

minimum of peak transmittance. All the TL measurements were performed using a linear heating rate of 5  C/s from RT up to 500  C, in a N2 atmosphere. Samples were irradiated at RT with a calibrated 90 Sr/90Y beta source with a dose rate of 0.1470 Gy/s. Four aliquots of powdered turquoise sample were used for each measurement. The incandescent background was directly subtracted from the TL data. 3. Results and discussion The studied turquoise (Fig. 1a) comes from Kathmandu (Tibet) as the typical tourist jewellery handcraft by local artisans using raw material from near outcrops. It has been analyzed to obtain its luminescent features and to evaluate its possible use as emergency dosimeter. By XRD analyses, the sample corresponds to turquoise phase with minor amounts of quartz [SiO2] and fluorapatite. A detailed study by ESEM showed the presence of the last two mineral phases together with monacite [(Ce,La,Nd,Th)PO4] and xenotime [YPO4] although the quantity of these minerals is not very abundant, only about 2%. Turquoise itself appears in three different morphologies (Fig. 1b,c), 1) a very small size of fibrous crystals forming the bulk matrix; 2) a medium size fibers filling fracture

Fig. 1. (a) Polished slide of the studied turquoise with darkseagreen2 RBG color (Red-180, Green-238, Blue-180). (b) Backscattered image of the sample under the ESEM microscope where vein-like areas can be observed. The displayed numbers correspond with the three described turquoise morphologies (see text). (c) Backscattered image under ESEM on the detail of the turquoise nodule morphology.

E. Crespo-Feo et al. / Radiation Measurements 45 (2010) 749–752

751

3

100

500

10000 3.4%

DTG

95

695

445 ºC

214 ºC

80 10.7%

75

Intensity (a.u.)

85

DTG (%/m)

TG (% weigth)

0 13.1%

90

7500

5000

435

2

310

618

285 657

TG

-2

350 ºC

1

2500

70 0

200

400

600

800

1000

200

Temperature (ºC)

300

400

500

600

700

800

Wavelength (nm)

Fig. 2. Thermogravimetric (TG) and Derivative Thermogravimetric (DTG) Analysis of the studied sample. Weight loss percentages and the corresponded temperatures are indicated.

zones; and 3) nodules of fibrous aggregates in the central parts of fracture zones. Each of them present small compositional variations in the amount of subtitutional cations in Cu2þ and Al3þ positions. As main features, Zn2þ content is 2.5%, 1.0%, and 0.4% in morphologies type 1, 2, and3, respectively, whereas Cr3þ presence is 0%, 0.01%, and 0.05%. In general, turquoise phase include REE such as Dy (460 ppm), Sm (170 ppm), Ce (110 ppm), Tb (80 ppm) and Nd (10 ppm). As could be expected by its composition, the DTA–TG curves show two endothermic peaks at 214 and 350  C that correspond to the H2O and OH-, respectively, released from the turquoise structure (Fig. 2). A third endothermic peak appears at 445  C, probably related to the decomposition of the anhydrous copper phosphate that forms an amorphous structure and have been confirmed by XRD analysis on the thermal analyzed residue. The natural TL glow curve of the powdered bulk turquoise (Fig. 3) displays two thermoluminescent peaks at 190  C and 330  C although the total intensity is not very high. The same aliquot after the TL analysis preheating and the additional irradiation shows no signal at all indicating that the previous observed signals are due to structural defects such as dehydration and dehydroxylation processes. This phenomenon was also observed by Murr (1979) who evidenced that irradiation effects or other alteration such as the loss of water

120

Fig. 4. X-Spatially-resolved Spectra cathodoluminescence under the Environmental Scanning Electron Microscope of the 3 different turquoise morphologies represented in Fig. 1b: (1) small fibrous crystals of the bulk matrix, (2) medium size fibers in fractures zones, and (3) nodules of fibrous aggregates in central parts of fracture zones.

altered the crystallinity of the turquoise. In any case, the low stability of the turquoise structure seems to avoid TL activation as a result of irradiation, and then its potential use as emergency dosimeter is not feasible. The spectrally-resolved cathodoluminescence has been performed on the three different turquoise morphologies (Fig. 1b). Three spectra are similar, with broad bands at w435 and w500 nm, and sharper ones at w310 and w657 nm, although big differences in CL intensity can be observed (Fig. 4). The spectral positions are almost the same except for the w500 nm band that shows a small shift towards higher wavelengths with the increasing intensity. The spectrum corresponding to morphology 3 shows two additional bands at w285 and w695 nm, this last one of high intensity (Fig. 4). Despite the high content in REE identified in the phosphate phases and in less amount in turquoise, no contribution to the CL emission seems to be present. Only part of the narrow bands could be associated to REE (e.g. w310 and w618 nm) although deeper studies have to be performed in order to establish a good correlation. In any case, Zn2þ must act as a quencher in the turquoise structure since a higher content in Zn2þ substituting for Cu2þ, lesser intensity of CL emissions. The well defined peak observed at w695 nm can be due to Cr3þ activation centers in the mineral lattice. Anyway, intrinsic defects seem to be the most important causes of turquoise luminescence since the three whole spectra are almost the same.

190ºC

Intensity (a.u.)

100

4. Summary

80 60

NTL ITL (1.3Gy) 330ºC

40 20 0 0

100

200

300

400

500

Temperature (ºC) Fig. 3. Thermoluminescence analysis of a natural (NTL) and a pre-irradiated turquoise aliquot (ITL) with 1.3 Gy.

The XRD and ESEM analyses of the studied turquoise show quartz as well as different phosphates such as fluorapatite, monacite and xenotime, all of them in accessorial amounts. Turquoise appears in three different morphologies, each of them with small compositional variations of substitutional cations in Cu2þ and Al3þ positions. TL spectrum seems to be related to structural defects such as dehydration and dehydroxylation processes since thermoluminescence peaks appears in temperatures similar to DTA–TG curves. This low stability of the turquoise structure avoids the TL activation as a result of irradiation reducing its potential use as emergency dosimeter. The CL spectra are similar for the three morphologies although the global intensity is quite different. Zn2þ must act as a quencher in the turquoise structure since a higher content in Zn2þ substituting for Cu2þ, lesser intensity of CL emissions. Narrow

752

E. Crespo-Feo et al. / Radiation Measurements 45 (2010) 749–752

bands at w310 and w618 nm could be associated to REE presence although peak observed at w695 nm is related to Cr3þ activation centers in the mineral lattice. In any case, intrinsic defects seem to be the most important causes of turquoise luminescence since the main broad bands at w435 and w500 nm are almost the same. Acknowledgments This work has been supported partly by a JAE-Doc CSIC postPhD. Contract and by the CICYT (FIS2007-61823) project. We are very grateful to the helpful comments of anonymous reviewers. References Belik, H.G., Çiftçi, M., 2008. Cathodoluminescence properties of turquoise. C.B.U. J. Sci. 4.2, 195–2000. Cid-Dresdner, H., 1965. The crystal structure of turquoise, CuAl6(PO4)4(OH)8$4H2O. Am. Mineral. 50, 283.

Clark, C.O., Poole, C.P., Farach, H.A., 1979. Variable-temperature electron spin resonance of turquoise. Am. Mineral. 64, 449–451. Diaz, J., Farach, H.A., Poole Jr., C.P., 1971. An electron spin resonance and optical study of turquoise. Am. Mineral. 56, 773–781. Frost, R.L., Reddy, B.J., Martens, W.N., Weier, M., 2006. The molecular structure of the phosphate mineral turquoise-a Raman spectroscopic study. J. Mol. Struct. 788, 224–231. Gaft, M., Reisfeld, R., Panczer, G., 2005. Luminescence Spectroscopy of Minerals and Materials, first ed. Springer, Berlin. ISBN-13-978-3-540-21818-7. Hunt, G.R., Salisbury, J.W., Lenhoff, C.J., 1972. Visible and near-infrared spectra of minerals and rocks: V. Halides, phosphates, arsenates, vanadates and borates. Mod. Geol. 3, 121–132. Kolitsch, U., Giester, G., 2000. The crystal structure of faustite and its copper analogue turquoise. Mineral. Mag. 64, 905–913. Murr, L.E., 1979. An electron microscopic study of crystalline turquoise. J. Mat. Sci. 14, 490–493. Reiche, I., Vignaud, C., Menu, M., 2000. Heat induced transformation of fossil mastodon ivory into turquoise ‘‘odontolite’’. Structural and elemental characterisation. Solid State Sci. 2, 625–636. Sharma, K.B.N., Moorthy, L.R., Reedy, B.J., Vedanand, S., 1988. EPR and electronic absorption spectra of copper bearing turquoise mineral. Phys. Lett. A 132, 293–297.