This article was published in an Elsevier journal. The attached copy is furnished to the author for non-commercial research and education use, including for instruction at the author’s institution, sharing with colleagues and providing to institution administration. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy
Radiation Measurements 42 (2007) 780 – 783 www.elsevier.com/locate/radmeas
Comparison of UV-IR radioluminescence and cathodoluminescence spectra of a potassium feldspar V. Correcher a,∗ , L. Sanchez-Muñoz a , J. Garcia-Guinea b , D. Benavente c , A. Delgado a a CIEMAT, Av. Complutense 22, Madrid 28040, Spain b Museo Nacional de Ciencias Naturales-CSIC, c/J. Gutierrez Abascal 2, Madrid 28006, Spain c Univ. Alicante, Apdo. 99, Alicante 03080, Spain
Received 19 December 2006; accepted 1 February 2007
Abstract This paper reports on the cathodoluminescence (CL) and radioluminescence (RL) emission properties of well-characterised K-rich feldspar (K0.98 Na0.02 Al1.02 Si2.98 O8 ). Both CL and RL spectra display three groups of components in the: (i) UV (290 nm); (ii) green (570 nm) and (iii) red-IR (640 nm) region. As observed in other aluminosilicates, the emission peaked at lower wavelengths (290 nm) can be associated with structural defects located in the twin-domain boundaries related to the recombination process in which the Na+ ion diffusion-limited is involved; green and red emissions are, respectively, associated with the presence of Mn2+ and Fe3+ ions. The ratio between the relative intensities peaked at 290 and 570 nm is about 3:7 for CL and 1:7 for RL; this fact indicates that the efficiency of recombination centres changes depending on the type of radiation, i.e., X-rays to obtain the RL spectra and electron beams for the CL emission. © 2007 Elsevier Ltd. All rights reserved. Keywords: Radioluminescence; Cathodoluminescence; Feldspar; Emission spectra
1. Introduction The luminescence properties of mineral phases, i.e., quartz and aluminosilicates, are of great interest to be employed in the field of retrospective dosimetry (Correcher et al., 1999), geological and archaeological dating (McKeever, 1985) and detection of food irradiation (Correcher et al., 1998). For these purposes the laboratory routine is mainly based on the analysis of the blue emission (at about 400 nm) of natural materials. Such emission is characterised by: (i) the sensitivity to radiation; (ii) high reproducibility; (iii) good dose linearity in the ranges of interest (up to 100 Gy) and (iv) high stability of the luminescence signal after long time of storage. This emission is produced by the incidence of the radiation on the aluminosilicates lattices that induces mobility of the alkalis causing a high number of electron–hole pairs in the lattice. Alkali self-diffusion through bulk and interfaces gives rise to a continuous alkali–oxygen bond splitting–linking processes with continuous formation–destruction of [AlO4 ]0 centres. Some ∗ Corresponding author. Fax: +34 1 346 6005.
E-mail address:
[email protected] (V. Correcher). 1350-4487/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.radmeas.2007.02.013
holes can be trapped, forming [AlO4 /M+ ] centres. When the supplied energy is enough, the recombination of the electrons with the hole trapped adjacent to Al-M+ reduces the presence of ionic charge compensators at the Al sites and induces the blue luminescence emission to [AlO4 ]0 centres (aluminium–hole centres) (Martini et al., 1994; Garcia-Guinea et al., 2001). However, there are some aluminosilicates which do not exhibit the 400 nm luminescence giving rise to potential errors in the estimation of the absorbed dose. In this sense, this paper reports about the physical parameters of the radioluminescence (RL) and cathodoluminescence (CL) emission of a well-characterised potassium rich feldspar from Madrid (Spain) where no blue emission has been detected. RL with X-rays is a sensitive method for obtaining information about the efficiency of recombination centres rather than shallow traps (Correcher and Garcia-Guinea, 2001). All the factors involved in the luminescence phenomena (i.e., lifetime, efficiency, emission spectra, etc.) depend directly on the crystalline phase which is mainly influenced by pressure and temperature. Thus, small variations in the lattice structure due to the presence of inclusions, impurities, substituted ions or
Author's personal copy
V. Correcher et al. / Radiation Measurements 42 (2007) 780 – 783
surface defects in ppm concentrations reveal changes in the intensity and wavelength position of the emission spectra. CL is a process whereby light is created from an energetic electron beam. CL supplies data about transient defects after irradiation on the surface of the lattice. CL is used in the identification of the migration and diffusion of some luminescent centres from the emission bands (Kalceff and Phillips, 1995).
781
50 Gy X-irradiated employing a Phillips MG MCN 101 X-ray tube with a current of 15 mA and a voltage of 25 kV delivering a dose rate of 10 Gy min−1 to the sample. CL experiments were performed using an electron beam with a current of 0.4 A and 10 kV of voltage at RT. Sample processing and measurements were made under red light to avoid the release of the trapped electrons from the semistable sites into hole centres (including luminescence centres) due to light sensitivity.
2. Experimental results and discussion 3. Results and discussion Measurements were carried out using a natural reddish opaque adularia sample (specimen LS176) collected from a hydrothermal fracture filling in El Verdugal granitic pegmatite from Colmenar Viejo (Madrid, Spain) (Sanchez-Munoz et al., 2006). The sample, with a mean molecular composition of orthoclase (Or), albite (Ab) and anorthite (An) of Or 97 Ab3 An0 , has been chemically analysed by X-ray fluorescence (XRF) using a PHILIPS PW-1404 with an Sc-Mo tube with analyser crystals of Ge, LIF220, LIF200, PE and TLAP and SuperQ manager from Panalytical-Spain as analytical software (Table 1). Pellets of 8 g of milled sample with 0.1 g of elbacite pressed under 20 TM and dried at 40 ◦ C in a climatic chamber have been employed. Non-destructive chemical analyses of major and minor elements were performed by electron probe microanalysis (EPMA) to provide information on the chemical homogeneity of the adularia crystal (Table 1). The sample was bound together with a polymer and softly polished offering a flat surface to the EPMA beam. The crystal-chemical characteristics of the feldspar were determined on data series of electron microprobe analyses (Jeol Superprobe JXA-8900M), bulk and channel-selected (TAP, PETJ, LIF, PETH) X-ray spectra search and by identification routines. The used standards were natural and synthetic crystals from the collection of the “Servicio de Microscopía Electrónica Lluis Bru”, Universidad Complutense de Madrid. The spot diameter of the probe was about 5 m and the operating conditions were 15 kV and 20 nA. Spectra were performed on cleaved chips of 3 × 3 × 2 mm3 (∼ 5 mg) of LS176 sample mounted with silicone oil onto aluminium disc using the spectrometer of Sussex University. The samples were not mechanically treated to avoid triboluminescence processes (Garcia-Guinea and Correcher, 2000). Signals were recorded over the 200–800 nm wavelength range, with a resolution of 5 nm for 100 point spectra, and 3 nm for 200 point spectra. All signals were corrected for the spectral response of the system. The RL was obtained during excitation of the samples with
Fig. 1 displays both RL and CL spectra obtained from the LS176 sample measured in the UV-IR region (from 200 to 800 nm). It is possible to appreciate how the use of different (i) sources of radiation (X-rays and electron beams) and (ii) doses does not induce significant changes in the spectral shape of the luminescence. However, one can observe how the behaviour of the samples (three replicates each), measured under the same conditions, are quite sensitive to radiation especially at higher wavelengths where the ratio between both areas of the curve (500–700 nm) is about 3:8. The variation of the intensity in the luminescence spectra can be due to defect creation by ionizing radiation of different energies. On the contrary, no significant differences (in intensity) could be noticed at higher energies (circa 290 nm). The experimental spectral data corresponding to CL and RL processes were fitted to three multiparameter Gaussian
Fig. 1. Comparison of UV-IR radioluminescence and cathodoluminescence spectra of the LS176 potassium rich feldspar measured under the same conditions. The ratio between the relative intensities, peaked at 290 and 570 nm is about 3:7 for CL and 1:7 for RL.
Table 1 Chemical composition of the analysed by electron probe microanalysis (EPMA) and X-ray fluorescence (XRF) Electron microprobe analysis Oxide %
SiO2 64.38
X-ray fluorescence Traces Rb Ppm 782
Al2 O3 18.89
Fe2 O3 0.03
Ba 76
Sr 17
MnO 0.03
CaO 0.01
Zr 16
Y 23
Na2 O 0.23
K2 O 16.72
P2 O 5 0.01
Ni 10
Co 4
Cr 12
Author's personal copy
782
V. Correcher et al. / Radiation Measurements 42 (2007) 780 – 783
Fig. 2. (a) RL emission spectrum of LS176 sample deconvoluted into three Gaussian components peaked at 293, 573 and 643 nm. Fitted Gaussian-peaks are represented with dashed lines; fitted solid line is calculated from the sum of the Gaussian peaks and dotted lines correspond to the experimental data; (b) second derivative of the RL experimental data to determine the local minima at peak positions.
Fig. 3. (a) CL emission spectrum of LS176 sample deconvoluted into three Gaussian components in the UV (at 296 nm), green (at 571 nm) and red (at 641 nm) region. Fitted Gaussian-peaks are represented with dashed lines; fitted solid line is calculated from the sum of the Gaussian peaks and dotted lines correspond to the experimental data; (b) second derivative of the CL experimental data to determine the local minima at peak positions.
functions using the Peak Fit program (supplied by Jandel Scientific Software). The minimum number of deconvoluted peaks that have reasonably been selected was three, taking into account both mathematical (second derivatives—Figs. 2b and 3b—, and the regression coefficient of fitting -r-) and physical meaning in each process. The smoothed second derivative of data determines the local minima at peak positions. Peaks with local maxima in the input data will produce second derivative local minima with values that fall below zero. Hidden peaks that evidence no local maxima in the input data will, if found by second derivatives, produce second derivative local minima with values that tend to zero. All the analysed parameters (i.e., position in nm, energy in eV, FWHM in nm, relative integrated area and relative intensity of the peaks) were refined to a confidence limit of 95% accuracy (Table 2). These data have been estimated from the experimental curves shown in Figs. 2a and 3a, where the dashed lines correspond to the calculated experimental dotted line that correspond to the experimental curve, for RL and CL luminescence spectra analyses. The three maxima appearing in both RL and CL processes exhibit similar energy positions of the emissions and FWHM. In addition, it is possible to appreciate how the behaviour of peak 3 seems to be independent of the type of radiation involved (Xray or electron beams); there are no changes in the estimated
parameters. However, a significant difference can be observed for the relative intensity of the analysed peaks. The ratio between peak 1 and 2 in the RL process is 1:7, whereas the same ratio in the CL process is 3:7. This behaviour involves different efficiency of recombination centres that changes depending on the type of radiation. The three dominants Gaussian-shaped curves peaked at about 290, 570 and 640 nm can be correlated with particular defect structures. The spectral data make possible to obtain information about the recombination site characteristic (Krbetschek et al., 1997). Thus, the 290 nm emission could be related to the defect-sites associated with the presence of Na in K aluminosilicate lattice (Garcia-Guinea et al., 1999). It has been seen in Na-rich feldspars (Ab) that this peak is the most important signal and is fitted Gaussian peaks, which make up the calculated fitted solid line. This encircling line is directly compared with the potentially useful for retrospective dosimetry (Correcher et al., 1999). Consequently, this waveband in K-feldspars could be used for dosimetric purposes. The green emission at can be attributed to Mn2+ substitutions in calcium sites in the lattice feldspars (Prescott and Fox, 1993). Finally, the red band at approximately 640 nm is produced when the irradiation reduces some Fe+3 into Fe+2 impurities (Kirsh and Townsend, 1988). Iron atoms can be substituted at a Si (tetrahedrally coordinated) or Al site in the aluminosilicate lattice,
Author's personal copy
V. Correcher et al. / Radiation Measurements 42 (2007) 780 – 783
783
Table 2 Physical parameters obtained from the deconvolution into Gaussian peaks of cathodoluminescence (CL) and radioluminescence (RL) of the LS176 K-rich feldspar (K0.98 Na0.02 Al1.02 Si2.98 O8 ) Peak 1
Peak 2
Peak 3
Radioluminescence
Peak position (nm) Peak energy (eV) Intensity of maxima (%) FWHM (nm) Area of peak (%) r
293 ± 3 4.23 ± 0.03 12.10 ± 0.8 36 ± 3 6.42 ± 0.5 0.992
0574 ± 9 2.16 ± 0.04 82.79 ± 8.2 69 ± 4 83.82 ± 5.3
643 1.93 5.03 132 9.76
± 11 ± 0.03 ± 0.4 ± 10 ± 0.6
Cathodoluminescence
Peak position (nm) Peak energy (eV) Intensity of maxima (%) FWHM (nm) Area of peak (%) r
296 ± 3 4.19 ± 0.03 28.29 ± 2.9 41 ± 4 18.54 ± 1.6 0.977
571 ± 8 2.17 ± 0.03 65.83 ± 5.9 67 ± 5 70.74 ± 6.5
641 ± 10 1.94 ± 0.04 6.08 ± 0.4 111 ± 10 10.73 ± 0.5
FWHM corresponds to Full Width at Half Maximum and r is the regression coefficient of fitting.
acting as recombination sites for holes or electrons, depending on the valence. Similar to the UV radiation in K-feldspars, X-rays could induce such a charge transfer between O2− and Fe3+ (Kirsh and Townsend, 1988). The probable reaction can be described as follows (Kirsh, 1991): Fe+3 + e− → Fe+2 Fe
+2
+
+ hole → [Fe
(irradiation step), +3 ∗
] → Fe
+3
+ (did-excitation step).
There is some important variation in this sample when compared with other K-rich aluminosilicates (Correcher and Garcia-Guinea, 2001). The sample here studied does not exhibit the typical UV-blue emission waveband that is usually the more important signal in many other specimens. Such a behaviour, not very common, could be linked to a low concentration of stable radiation-induced defects in the lattice and could induce potential mistakes in the dose estimation when routine laboratory tests are applied for dating or retrospective purposes. 4. Conclusions The well-characterised K-rich feldspar here studied under RL and CL, exhibits three emission bands peaked at 290, 570 and 640 nm that can be associated, respectively, with the presence of Na+ , Mn2+ and Fe3+ . The correlation of the RL and CL emission wavebands to the aforementioned cations has been possible using chemical analysis by means of XRF and EPMA. Unlike many other aluminosilicates, this sample lack of the UV-blue emission. The similar CL and RL spectra (in shape) indicate the common origin of the emission centres, regardless of the type of radiation; however, the different intensity of luminescence spectra in the green and red band could be due to variations in the efficiency of recombination centres that depends on the type of radiation based on defect creation. Acknowledgements We are grateful to Prof. Dr. P.D. Townsend for the analyses in the high sensitivity spectrometer of Sussex (UK). This work
has been supported by the C.I.C.Y.T. CGL2004-03564/BTE and Comunidad Autonoma de Madrid (CAM) MATERNAS (S0505/MAT/0094) projects. References Correcher, V., Garcia-Guinea, J., 2001. On the luminescence properties of adularia feldspar. J. Lumin. 93 (4), 303–312. Correcher, V., Muniz, J.L., Gomez-Ros, J.M., 1998. Dose dependence and fading effect of the thermoluminescence intensities in gamma-irradiated paprika. J. Sci. Food Agric. 76, 149–155. Correcher, V., Gomez-Ros, J.M., Delgado, A., 1999. The use of albite as a dosemeter in accident dose reconstruction. Radiat. Prot. Dosim. 84 (1–4), 547–549. Garcia-Guinea, J., Correcher, V., 2000. Luminescence spectra of alkali feldspars: influence of crushing on the ultraviolet emission band. Spectrosc. Lett. 33, 103–113. Garcia-Guinea, J., Townsend, P.D., Sanchez-Munoz, L., Rojo, J.M., 1999. Ultraviolet-blue ionic luminescence of alkali feldspars from bulk and interfaces. Phys. Chem. Miner. 29, 658–667. Garcia-Guinea, J., Correcher, V., Rodriguez-Badiola, E., 2001. Analysis of luminescence spectra of leucite (KAlSiO4 ). Analyst 126, 911–916. Kalceff, M.A.S., Phillips, M.R., 1995. Cathodoluminescence microcharacterization of the defect structure of quartz. Phys. Rev. B 52 (5), 3122–3134. Kirsh, Y., 1991. The kinetics of the red and blue bands in the phosphorescence of X-irradiated albite. Phys. Status Solidi (A) 127, 265–273. Kirsh, Y., Townsend, P.D., 1988. Speculations on the blue and red bands in the tl emission-spectrum of albite and microcline. Nucl. Tracks. Radiat. Meas. 14 (1–2), 43–49. Krbetschek, M.R., Gotze, J., Dietrich, A., Trautmann, T., 1997. Spectral information from minerals relevant for luminescence dating. Radiat. Meas. 27, 695–748. Martini, M., Spinolo, G., Vedda, A., Arena, C., 1994. Phosphorescence and thermally stimulated luminescence of amorphous SiO2 . Solid State Commun. 91 (9), 751–756. McKeever, S.W.S., 1985. Thermoluminescence of Solids. Cambridge University Press, New York. Prescott, J.R., Fox, P.J., 1993. 3-dimensional thermoluminescence spectra of feldspars. J. Phys. D 26, 2245–2254. Sanchez-Munoz, L., Garcia-Guinea, J., Sanz, J., Correcher, V., Delgado, A., 2006. Ultraviolet luminescence from defect complexes in the twin boundaries of K-feldspar. Chem. Mater. 18, 3336–3342.
