Physica B 406 (2011) 3256–3260
Contents lists available at ScienceDirect
Physica B journal homepage: www.elsevier.com/locate/physb
Intense ultraviolet and blue upconversion emissions in Yb3 þ –Tm3 þ codoped stoichiometric Y7O6F9 powder Mo Ma a,b, Changfu Xu a,b,n, Liwen Yang a,b, Guozhong Ren a,b, Jianguo Lin a,b, Qibin Yang a,b a b
Key Laboratory of Low Dimensional Materials and Application Technology, Ministry of Education, Xiangtan University, Xiangtan 411105, China Institute of Modern Physics, Faculty of Material and Photoelectronic Physics, Xiangtan University, Xiangtan 411105, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 November 2010 Received in revised form 6 May 2011 Accepted 18 May 2011 Available online 24 May 2011
Stoichiometric Y7O6F9 powder codoped with Yb3 þ –Tm3 þ was synthesized via co-precipitation and subsequent calcining route. The results of X-ray diffraction and transmission electron microscopy reveal that when the calcining temperature is beyond 800 1C, orthorhombic YF3 nanoparticles can be completely oxidized into orthorhombic Y7O6F9 powder. Under the excitation of a 980 nm laser, Y7O6F9 powder exhibits multicolor UC emission in regions spanning the UV to the NIR. In addition, the upconversion emission intensities of YF3, Y7O6F9 and Y2O3 powders were compared under the same dopant condition (Yb/Tm ¼ 5/0.5 mol%). The low phonon energy revealed by Raman spectra helped to understand the high efﬁcient upconversion emission of Y7O6F9 and the main phonon vibration of Y7O6F9 lies at 472 cm 1, which is far lower that of Y2O3 (at 708 cm 1). Our results indicate that orthorhombic rare earth ions doped Y7O6F9 is an efﬁcient matrix for UV and blue UC emission, and has potential applications in color displays, anti-counterfeiting and multicolor ﬂuorescent labels. & 2011 Elsevier B.V. All rights reserved.
Keywords: Oxyﬂuoride Optical material Upconversion Y7O6F9 Raman
1. Introduction Rare earth (RE) ions doped upconversion (UC) materials attract more and more attention because of their potential use in optical communications, solid-state lasers [1–4], solar cells [5,6], photocatalysis [7–10] and photoelectronics [11,12]. Especially in the latest decade, the application of UC materials in bioimaging and biomedicine has progressed rapidly [13,14]. Many host materials such as NaYF4 , NaYbF4 , Y2O3  and YF3  can realize a high efﬁcient multicolor UC emission under the near-infrared (NIR) laser excitation. Amongst these materials YF3 is one of the most promising matrices due to its good optical transparency in a wide wavelength range and the minimization of the excited state quenching of the rare earth ions [19–21], especially its low phonon energy. In comparison ﬂuorides usually exhibit low phonon energies ( 350 cm 1), but the phonon energies of oxides are relatively high and generally larger than 500 cm 1. For example, the maximum phonon energy of YF3 lattice is below 500 cm 1  and the intrinsic phonons of Y2O3 are about 600 cm 1 . It is obvious that ﬂuorides have lower phonon energy than oxides. In addition, YF3 is chemically and thermally unstable so that it can be easily oxidized into yttrium oxyﬂuoride under high temperature in air. RE
oxyﬂuorides combine the advantages of both ﬂuoride and oxide crystals, so they can show better chemical and thermal stability than the ﬂuorides and lower phonon energy than the oxides and will be desired UC host materials. Uptil now by controlling the dopant–host combination and dopant concentration, multicolor UC emissions from infrared to deep ultraviolet including white emission have been realized in oxyﬂuorides glass and ceramics [24,25]. However, there are few reports about UC emission in stoichiometric RE oxyﬂuoride. Zhang’s report had proved stoichiometric oxyﬂuoride Y6O5F8 to be a nice host material for upconversion . YOF had been systematically and broadly researched for its nice photoluminescence and chemical stability . Y7O6F9, which is one of the important RE oxyﬂuoride, and of the end-centered orthorhombic structures have not been reported about its upconversion. In this investigation stoichiometric Y7O6F9 powder codoped with Yb3 þ –Tm3 þ exhibiting intense ultraviolet (UV) and blue UC emissions under the 980 nm laser excitation was reported. In addition, YF3 and Y2O3 powders with same doping condition were prepared to compare their upconversion property with Y7O6F9 powder and the Tm doping concentration effect on the upconversion intensity was also investigated.
2. Experimental n
Corresponding author at: Key Laboratory of Low Dimensional Materials and Application Technology, Ministry of Education, Xiangtan 411105, China. Tel.: þ 86 732 58292195. E-mail address: [email protected]
(C. Xu). 0921-4526/$ - see front matter & 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.physb.2011.05.035
Stoichiometric Y7O6F9 powder codoped with Yb3 þ –Tm3 þ was synthesized via a co-precipitation and subsequent calcining route. All reagents of analytical grade were purchased from Sinapharm
M. Ma et al. / Physica B 406 (2011) 3256–3260
Chemical Reagent Co. Ltd. In a typical procedure, 2 mmol RE(NO3)3 including 1.89 mmol Y3 þ , 0.1 mmol Yb3 þ and 0.01 mmol Tm3 þ was ﬁrst dissolved in 30 ml deionized water. Then 1 ml hydroﬂuoride acid (40%) was added dropwise into the above solution at room temperature under vigorous stirring to form the mushy gelation. This gelatin was unstable when it was heated at 50 1C. After an hour, white precipitate was collected from the unstable gelatin by centrifugation and then was rinsed with deionized water and dried at 60 1C for 12 h in vacuum. Finally the precipitate was calcined in air at different temperature for 6 h to obtain the samples. If the precipitate was calcined in N2 atmosphere, YF3 was obtained for comparison. In addition, another sample Y2O3 power with the same doping condition was prepared with the following urea homogeneous precipitation method . Lanthanide nitrate solution containing 1.89 mmol Y(NO3)3, 0.1 mmol Yb(NO3)3 and 0.01 mmol Tm(NO3)3, and 50 ml urea (1 M) were mixed together and stirred vigorously to form uniform mixture solution. The mixture solution was heated to 85 1C with stirring for 4 h. The precipitation was obtained by ﬁltering and rinsed for three times with deionized water, then dried at 70 1C for 12 h. Finally the obtained precursor powder was calcined at 800 1C for 5 h to get Y2O3 powders. The crystal structures of the as-prepared samples were determined by X-ray diffraction (XRD) using the copper Ka radiation. The morphologies and the structures of the samples were characterized using transmission electron microscopy (TEM, JEM2100) and selected area electron diffraction (SAED). UC spectra were recorded on a spectrophotometer (R-500) under excitation by a 980 nm laser diode (LD). The dependence of the UC emission intensities on pumping power densities was obtained by changing the excitation power. Raman spectra were measured with R-500 spectrometer with the excitation of 531.9 nm laser. All the measurements were performed at room temperature.
3. Results and discussion Fig. 1 shows the XRD patterns of the as-prepared samples calcined at different temperatures. For the uncalcined sample, broader diffraction peaks can be observed (see Fig. 1a). Compared to the standard PDF data the XRD peaks are well in accordance with that of orthorhombic YF3 (JCPDS 32-1431). Based on the Sherrer’s equation (D ¼ 0:89l=b cos y, where D is the average crystal size, l is the X-ray wavelength of 1.5406 nm, y and b are the diffraction angle
Fig. 1. XRD patterns of YF3 nanoparticles uncalcined and calcined in air at different temperature: (a) uncalcined; (b), (c), (d), (e), calcined at 400, 500, 700 and 800 1C, respectively. Pure Y7O6F9 powder can be obtained at 800 1C.
and the full-width at half-maximum of the considered peak), the grain size of YF3 is about 9 nm. With the increasing of calcining temperature the peaks of YF3 become sharper, implying that the grain size of YF3 is growing up (Fig. 1b). No phase transition occurs until the calcining temperature is beyond 500 1C. From the XRD results of the sample calcined at 500 1C shown in Fig. 1c it can be seen that the peak at 27.681 becomes abnormally broader and a shoulder at 28.141 appears. It means that partial F ions in YF3 are oxidized by O2 and one type of oxyﬂuoride is formed. According to the XRD results of the samples calcined at 700 1C (Fig. 1d) it can be inferred that this oxyﬂuoride might be Y7O6F9 (JCPDS 80-1126). With further increasing of calcining temperature the oxidation of YF3 becomes more complete, and it is hard to ﬁnd the peaks of YF3 when the calcining temperature is beyond 800 1C. As shown in Fig. 1e all XRD peaks can be well indexed to the stoichiometric orthorhombic Y7O6F9. The morphologies and the microstructures in the transition from YF3 to Y7O6F9 were further investigated by TEM and HRTEM. Fig. 2a depicts the typical TEM image of the uncalcined YF3. It can be observed that YF3 particles are spindle-shaped with their average size of about 500 nm length and 200 nm width. The corresponding HRTEM image (see Fig. 2a0 ) reveals that these spindle-shape YF3 particles consist of a large number of nanocrystals. The typical interplanar spacing of these nanocrystals is ˚ which is matched with the (2 1 0) lattice planes of about 2.85 A, the orthorhombic YF3. Its value is little smaller than that of ˚ for the radius of Yb3 þ (0.858 A) ˚ is smaller standard YF3 (2.88 A) ˚ and Yb3 þ was doped in YF3 nanocrysthan that of Y3 þ (0.893 A) tals. The morphology and the size of the samples have no obvious changes (see Fig. 2b, c) when the calcined temperature below 700 1C, but the size of YF3 nanocrystals becomes large together with the crystal defects such as the vacancies and dislocations decreasing quickly. Especially, the (1 1 1) interplanar spacing of YF3 nanocrystals decreases. For the sample calcined at 400 1C its ˚ while for the sample at 500 1C, it value is 3.11 A˚ value is 3.19 A, (see Fig. 2b0 , c0 ). This can be attributed to the partial oxidation of YF3 nanoparticles to form Y7O6F9. The interplanar spacing of the (1 7 1) lattice planes in orthorhombic Y7O6F9 is 3.12 A˚ (see Fig. 2d0 ), smaller than that of the (1 1 1) lattice planes in orthorhombic YF3. Fig. 2e, e0 shows typical TEM and HRTEM images of pure Y7O6F9 power. The plane spacings in the HRTEM ˚ The fast image (see Fig. 2e0 ) were measured to be 5.36 and 5.27 A. Fourier transform (FFT) of the HRTEM image reveals that the angle between these lattice planes is 921, which is in accordance with the theoretical value. According to the XRD results these lattice planes can be well indexed as (1 2 0) and ð0 1 1Þ of orthorhombic Y7O6F9, respectively. YF3 and Y2O3 under the same doping condition were prepared. Fig. 3 shows XRD patterns of other samples. Each peaks of these two samples can be indexed to YF3 (Fig. 3a, JCPDS 32-1431) and Y2O3 (Fig. 3b, JCPDS 41-1105) and no impurity peak can be found. Using the strongest diffraction peaks at 2y ¼27.461 for YF3, 2y ¼28.191 for Y7O6F9 and 2y ¼29.161 for Y2O3, the mean sizes of the powders were estimated to be 29, 47 and 28 nm, respectively. Raman spectra can be used to characterize their phonon energy. Fig. 4 displays the Stokes Raman shifts of (a) Y7O6F9, (b) YF3 and (c) Y2O3 with the same doping condition (Yb/Tm¼ 5/0.5 at%) at room temperature (lex ¼531.9 nm). Under the excitation of 531.9 nm laser, several peaks of Raman shifts can be observed for each sample in the range of 150–1000 cm 1. The peaks for Y7O6F9 appear at 223, 351, 412, 472, 593, and 809 cm 1. Six peaks in 184, 270, 354, 432, 504 and 604 cm 1 exist for YF3. More phonon spectra can be observed for Y2O3 and exist in 203, 335, 441, 479, 528, 569, 618, 708, 802, and 959 cm 1, which is in agreement with Rai’s report . The main phonon vibration of
M. Ma et al. / Physica B 406 (2011) 3256–3260
Fig. 2. TEM images of YF3 nanoparticles uncalcined and calcined at different temperatures for 6 h. Upper was the appearance ((a) uncalcined; (b) 400 1C; (c) 500 1C; (d) 700 1C; (e) 800 1C) and the lower was the corresponding high resolution TEM image ((a0 ) uncalcined; (b0 ) 400 1C; (c0 ) 500 1C; (d0 ) 700 1C; (e0 ) 800 1C), the inset between (e) and (e0 ) was the fast fourior transform (FFT) of HRTEM image in (e0 ).
Fig. 3. XRD patterns of YF3 (a) and Y2O3 (b) powders.
Fig. 4. Raman spectra of (a) Y7O6F9, (b) YF3 and (c) Y2O3 with the same doping condition (Yb/Tm ¼5/0.5 at%) at room temperature (lex ¼ 531.9 nm).
Fig. 5. UC emission of YF3 (a), Y7O6F9(b) and Y2O3 (c) powers codoped with Tm3 þ /Yb3 þ (0.005:0.05 mol%).
three samples can be found in 472 cm 1 for Y7O6F9, 184 cm 1 for YF3 and 708 cm 1 for Y2O3, respectively. Luminescence efﬁciency is decided by the nonradiative process of the host materials. High phonon vibration will increase nonradiative relaxation rate and hence decrease the luminescent efﬁciency. Raman spectra reveal the main vibration modes of three samples, which can be used to estimate the comparative emission intensity of these three samples. Fig. 5 is the UC spectra of Yb3 þ –Tm3 þ (0.005:0.05% mol) codoped YF3 (a), Y7O6F9 (b) and Y2O3 powders under the excitation of a 980 nm LD with a pump density of 1 W/cm2. Both YF3 and Y7O6F9 powders have multicolor UC emission spanning the UV to the NIR. It mainly includes three UV emission bands at 291, 353 and 362 nm, respectively, two blue bands centered at 447 and 477 nm, one green band at 512 nm, one yellow band at 580 nm, two red bands at 647 and 692 nm, and one strong NIR band centered at 808 nm. These UC bands correspond to the 3 P0-3H6, 3P0-3F4, 1D2-3H6, 1D2-3F4, 1D2-3H5, 1G4-3H6, 1 D2-3H4, 3F2/3F3-3H6 and 3H4-3H6 transitions of Tm3 þ ions, respectively. Especially one can see that the intensities of UV and blue UC emission are quite intense compared to that of NIR
M. Ma et al. / Physica B 406 (2011) 3256–3260
emission at 808 nm, implying YF3 and Y7O6F9 is an efﬁcient matrix for ultraviolet and blue UC emissions. Compared with Y7O6F9, the emission intensity of YF3 increased a little but the ultraviolet intensity of YF3 is obviously stronger than that of Y7O6F9. YF3 has a lower intrinsic phonon energy than Y7O6F9, but the size of YF3 is far smaller than that of Y7O6F9, hence the emission intensity of YF3 is comparable with that of Y7O6F9 because of the size effect . However, the emission intensity of Y2O3 is comparably weak and the ultraviolet emission band can hardly be observed. Though the size has effect on the emission intensity, the emission efﬁciency of Y7O6F9 powder is obviously better than that of Y2O3 powder. Under the same doping condition, the emission intensity of Y7O6F9 powder is little weaker than that of YF3, but much stronger than that of Y2O3. According to these results it is believed that Y7O6F9 is a nice host matrix for upconversion. Y7O6F9 powders at different doping condition (Yb:Tm¼ 5:x mol%, x ¼1, 0.5, 0.25, 0.1) all show intense upconversion emission, and the upconversion spectra are shown in Fig. 6a. It is noticed that the changes are different for the emission intensity of each emission band with different Tm content. Fig. 6b shows
Fig. 7. Double logarithm plots of emission intensity vs. excitation power of Tm3 þ /Yb3 þ codoped Y7O6F9 powder.
the relationship between the integrated intensity of main emission bands and Tm content. When Yb/Tm doping content is 5/0.25 (mol%) strongest emission in each band can be obtained. According to the energy level diagrams of Tm3 þ and Yb3 þ ions as shown in Fig. 7 and other’s research results , the possible UC mechanisms to populate the excited levels might be realized via the multiple phonon-assisted ET processes from Yb3 þ ions to Tm3 þ ions as follows: 2 F5/2(Yb3 þ ) þ 3H6(Tm3 þ )-3H5 (Tm3 þ )þ 2F7/2(Yb3 þ )—— I 3F4(Tm3 þ )þmultiphonon relaxation 2
F5/2(Yb3 þ ) þ 3F4(Tm3 þ )-3F2/3F3(Tm3 þ )þ 2F7/2(Yb3 þ )—— II 3H4(Tm3 þ )þmultiphonon relaxation
F5/2(Yb3 þ ) þ 3H4(Tm3 þ )-1G4(Tm3 þ )þ 2F7/2(Yb3 þ )——III
F5/2(Yb3 þ ) þ 1G4(Tm3 þ )-1D2(Tm3 þ )þ 2F7/2(Yb3 þ )——IV
F5/2(Yb3 þ ) þ 1D2(Tm3 þ )-3P1,2(Tm3 þ )þ 2F7/2(Yb3 þ )—— V 3P0(Tm3 þ )þ multiphonon relaxation Usually the phonon-assisted ET rate (g) can be expressed as g ¼Pe aDE, where P is the probability of the ET process without
Fig. 6. (a) Upconversion spectra of Yb/Tm codoped Y7O6F9 (Yb/Tm ¼5/xmol%, x¼ 1, 0.5, 0.25, 0.1) powders under different Tm contents; (b) main emission band integrated intensity vs. Tm content.
energy mismatch, DE is the mismatch for the energy transfer, and R is the electron-lattice coupling coefﬁcient. Considering the mismatch of energy between the 2F5/2-2F7/2 (Yb3 þ ) and 1 G4-1D2 (Tm3 þ ) transitions (IV) is greater than 3000 cm 1, this ﬁnal transfer should occur with little probability. As a result the intensity of this peak at 453 nm would be extremely weak, which is against our experimental results. Therefore there should exist other UC process. Here we consider that the most likely transition is the nearly resonant cross relaxation energy transfer (CRET) betweens Tm3 þ ions: 3F2 þ 3H4-3H6 þ 1D2 (CRET), which is also a four-photon UC process to populate the excited 1D2 level [31,32]. For the better understanding of luminescent dynamics and UC mechanism, the power dependent UC behaviors of the observed bands are investigated. Generally, for unsaturated UC processes the UC luminescence intensity (Iup) is related to the pump n infrared one (IIR) via the formula, IUP pIIR , where n is the number of pump photons required to populate the upper emitting level and its value can be obtained from the slope of the line in the plot n of log IUP vs. log IIR . Fig. 8 shows the log–log plot of the UC emission intensity vs. the pumping power in Yb3 þ –Tm3 þ codoped orthorhombic Y7O6F9. Slope values of the linear, which ﬁts with the experimental data in the range of low pump power are 1.61, 2.61, 2.67, 3.74 and 4.44 for the UC bands at 808, 647, 477, 362 and 347 nm, respectively. The results indicate that two pump photons are necessary to produce the UC emissions centered at 808 nm, three photons to produce the ones at 647 and 477 nm, four pump photons to produce the UC ones at 453
M. Ma et al. / Physica B 406 (2011) 3256–3260
Acknowledgment This work was supported by Hunan Education Department Foundation of China (Grant no. 06C818). References
Fig. 8. Population and photoluminescent processes.
and 362 nm, and ﬁve pump photons to produce the UC emissions at 347 and 291 nm. The results are consistent with other previous reports in Yb3 þ –Tm3 þ codoped YF3 and NaYF4 [33,34].
4. Conclusion In summary, stoichiometric orthorhombic Y7O6F9 powder codoped withYb3 þ –Tm3 þ was synthesized via co-precipitation and subsequent calcining route. When excited by a 980 nm laser the oxyﬂuoride powder exhibits intense multicolor UC emission originated from 4 f–4 f transitions of Tm3 þ ions, including UV, blue, green, yellow, red and NIR. The intensities of UV and blue UC emission are comparable to that of NIR one, implying Y7O6F9 is an efﬁcient matrix for UV and blue UC emission. Under the same doping condition the emission intensity of Y7O6F9 powder is little weaker than that of YF3, but much stronger than that of Y2O3. The main phonon vibration of Y7O6F9 (472 cm 1) revealed by Raman spectra is far lower than the phonon energy of Y2O3 (472 cm 1), which helps to understand the high efﬁcient upconversion emission of Y7O6F9. The power dependence studies indicate that multiple phonon-assisted ET processes from Yb3 þ to Tm3 þ ions are responsible for these UC emissions. Our results indicate that the RE ions doped orthorhombic Y7O6F9 powder will have potential applications in color displays, anti-counterfeiting and multicolor ﬂuorescent labels.
 E. Downing, L. Hesselink, J. Ralston, R. Macfarlane, Science 273 (1996) 1185.  M. Godlewski, S. Yatsunenko, A. Nadolska, A. Opalin´ska, W. Łojkowski, K. Drozdowicz-Tomsia, E.M. Goldys, Opt. Mater. 31 (2009) 490.  K. Binnemans, Chem. Rev. 109 (2009) 4283.  S. Sanders, R.G. Waarts, D.G. Mehuys, D.F. Wetch, Appl. Phys. Lett. 67 (1995) 1815. ¨  B. Ahrens, P. Loper, J.C. Goldschmidt, S. Glunz, B. Henke, P.T. Miclea, S. Schweizer, Phys. Status Solidi (a) 12 (2008) 2822. ¨ ¨  A. Shalav, B.S. Ridchards, T. Trupke, K.W. Kramer, H.U. Gudel, Appl. Phys. Lett. 86 (2005) 013505.  J. Wang, G. Zhang, Z.H. Zhang, X.D. Zhang, G. Zhao, et al., Water Res. 40 (2006) 2143.  J. Wang, F.Y. Wen, Z.H. Zhang, X.D. Zhang, et al., J. Photochem. Photobiol.(A) 180 (2006) 189.  K.B. Zhou, X. Wang, X.M. Sun, Q. Peng, Y.D. Li, J. Catal. 229 (2005) 206.  X. Wang, J. Zhuang, Q. Peng, Y.D. Li, Inorg. Chem. 45 (2006) 6661.  K. Kawano, K. Arai, H. Yamada, N. Hashimoto, R. Nakata, Sol. Energy Mater. Sol. Cells 48 (1997) 35.  M. Godlewski, A. Yatsunenko, A. Nadolska, A. Opalin´ska, W. Łojkowski, K.D. Tomsia, E.M. Goldys, Opt. Mater. 31 (2009) 490.  M. Huang, S. Mao, H. Feick, H. Yan, Y. Wu, H. Kind, E. Weber, R. Russo, P. Yang, Science 292 (2001) 1897.  L. Xiong, Z. Chen, M. Yu, F. Li, C. Liu, C. Huang, Biomaterials 30 (2009) 5592.  S. Zeng, G. Ren, Q. Yang, J. Alloys Compd. 493 (2010) 476.  S. Zeng, G. Ren, W. Li, C. Xu, Q. Yang, J. Phys. Chem. C 114 (2010) 10750.  C.B. Zheng, Y.Q. Xia, F. Qin, Y. Yu, J.P. Miao, Z.G. Zhang, W.W. Cao, Chem. Phys. Lett. 496 (2010) 316.  G. Wang, W. Qin, J. Zhang, J. Zhang, W. Yan, C. Cao, L. Wang, G. Wei, P. Zhu, R. Kim, J. Phys. Chem. C 112 (2008) 12161.  I. Iparraguirre, J. Azkargorta, R. Balda, J. Ferna´dez, Opt. Mater. 27 (2005) 1697.  V. Trnovcova´, L.S. Garashina, A. sˇ kubla, P.P. Fedorov, R. Cicka, E.A. Krivandina, B.P. Sobolev, Solid State Ionics 157 (2003) 195.  M. Nikl, A. Yoshikawa, A. Vedda, T. Fukuda, J. Cryst. Growth 292 (2006) 416.  H.E. Rast, H.H. Caspers, S.A. Miller, Phys. Rev. 180 (1969) 890.  J.A. Capobianco, J.C. Boyer, F. Vetrone, A. Speghini, M. Bettinelli, Chem. Mater. 14 (2002) 2915.  L. Feng, B. Lai, J. Wang, G. Du, Q. Su, J. Lumin. 130 (2010) 2418.  T. Pang, W. Cao, M. Xing, W. Feng, S. Xu, Physica B 405 (2010) 2216.  S. Wang, R. Deng, H. Guo, S. Song, F. Cao, X. Li, S. Su, H. Zhang, Dalton Trans. 39 (2010) 9153.  Z. Li, L. Zhang, L. Zhang, J. lumin. 126 (2007) 481.  J. Silver, M.I. Maritinez-Rubio, T.G. Ireland, G.R. Fem, R. Withnall, J. Phys. Chem. B 105 (2001) 948.  K. Mishra, N.K. Giri, S.B. Rai, Applied Physics B DOI 10.1007/s00340-01 1-4379-5.  G.S. Yi, H.C. Lu, S.Y. Zhao, Y. Ge, W.J. Yang, D.P. Chen, L.H. Guo, Nano lett. 4 (2004) 2191. ¨ zen, F. Pelle´, J. Lumin. 60/61 (1994) 212 D.  X. Wu, J.P. Denis, G. O  Y. Chen, Y. Wang, P. Yu, Huang, Appl. Phys. Lett. 91 (2007) 051920.  G.F. Wang, W.P. Qin, G.D. Wei, L.L. Wang, et al., J. Fluorine Chem. 130 (2009) 158.  X. Chen, W. Wang, X.Y. Chen, J. Bi, L. Wu, Z. Li, X. Fu, Mater. Lett. 63 (2009) 1023.