Investigations on photoluminescence and cathodoluminescence properties of Ca3La6(SiO4)6:Tb3 +, Mn2 +

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 165 (2016) 85–89

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

Investigations on photoluminescence and cathodoluminescence properties of Ca3La6(SiO4)6:Tb3 +, Mn2 + Jia Zhang ⁎, Beibei Zhou, Xichen Wang School of Physics and Electronic Electrical Engineering, Huaiyin Normal University, 111 West Chang Jiang Road, Huai'an 223001, China

a r t i c l e

i n f o

Article history: Received 12 February 2015 Received in revised form 16 July 2015 Accepted 8 April 2016 Available online 12 April 2016 Keywords: Phosphor Luminescence Ca3La6(SiO4)6:Tb3+, Mn2+

a b s t r a c t Tb3+/Mn2+ activated Ca3La6(SiO4)6 (CLS) phosphors were prepared by solid-state reaction method, and their photoluminescence and cathodoluminescence (CL) properties were investigated. The CLS:Tb3+ sample shows a yellowish green emission under 377 nm excitation, and the excitation spectrum reveals the excitation peaks between 340 and 390 nm can match with the near-ultraviolet LED chip. Excellent thermal stability has been obtained in the CLS:Tb3+ phosphor by studying the temperature dependence of the Tb3+ emission intensity. By introducing Mn2+ into CLS:Tb3+, tunable emissions are generated due to the efficient energy transfer from Tb3+ to Mn2+. The CL spectrum of CLS:Tb3+ displays that the characteristic 5D4-7FJ (J = 6 − 3) transitions of Tb3+ are found under electron beam excitation. The above investigation results imply that the CLS:Tb3+, Mn2+ phosphors could have potential applications on LEDs and FEDs. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In recent years, luminescence materials are of crucial importance to display and lighting field, and have been developed in the light-emitting diodes (LEDs), field emission displays (FEDs), X-ray imaging scintillators, infrared quantum counter detectors and biomarkers [1–3]. Especially, the high efficiency phosphor converted LEDs (pcLED) lamps for general lighting could significantly reduce lighting energy consumption and do not need any harmful ingredients in comparison with conventional light sources [4,5]. Thus, white LEDs with high color rendering index and sufficient color reproducibility have attracted considerable interest at present. The most common and simplest commercial strategy to achieve white light is the combination of a blue LED chip and the yellow-emitting YAG:Ce3 + phosphor [6]. However, this approach causes several serious problems, such as low color rendering index (CRI) and high correlated color temperature (CCT) due to the deficiency of a red fluorescent component [7]. To resolve these problems, an alternative approach was developed, which involves the manufacture of near-UV LED chips by blending red-, green-, and blue-emitting phosphors together on top of an n-UV LED chip to assemble WLEDs [8]. The white light generated by this method is highly favored due to the high CRI and controllable CCT [9]. Since the performances of white LEDs strongly depend on the luminescence characters of phosphors used, it is important to explore new phosphors with excellent luminescence properties. Moreover, it is also interesting to synthesis novel multicolor phosphors, which can overcome the shortages of phosphors combination, such as different degradation rates and re-absorption ⁎ Corresponding author. E-mail address: [email protected] (J. Zhang).

http://dx.doi.org/10.1016/j.saa.2016.04.029 1386-1425/© 2016 Elsevier B.V. All rights reserved.

between phosphors [2]. And, the realization of tunable multicolor emission under a single excitation wavelength in phosphors is beneficial for the possible application in display device and multiplexed biological detection [10]. Generally, tunable emission can be obtained by employing the energy transfer (ET) between the luminescent ions, for instance, CaAl2Si2O8:Eu2+, Mn2+, Ca3Sc2Si3O12:Ce3+, Tb3+, and Ca2Al3O6F:Ce3+, Tb3+ [11–13]. For the time being, phosphors of this kind are also urgently needed. FEDs have been also regarded as one of the most promising next generation flat panel displays due to the advantages of wide viewing, quick response time, good contrast ratio. Compared with the cathoderay tubes, the FEDs operate at lower voltage (≤10 kV) and higher current density (10–100 A/cm2), so the phosphors used should have high efficiency at low voltages, high resistance to current saturation and long service time [14]. As present, the phosphors of high luminous efficiency for FEDs are most sulfide-based luminescent materials, such as ZnS:Cu, Al and SrGa2S4:Ce [15]. However, they show low stabilities under the high-energy electron bombardment and are not environmentally friendly, which seriously prohibit their use [16]. Hence, more and more investigations have been focused on the oxide-based phosphors due to their higher stability and environmental friendliness, which are still urgently required currently. Rare earth (RE) ions are known to have been playing an important role in modern lighting and display field, and their unique electronic structures enable RE ions in solids to emit photons efficiently in the spectral region from UV to visible light [17]. Tb3+ is known to be an important green light emitting luminescent activator which has the main 7 FJ, 5D3, and 5D4 states. Mn2 + activated phosphors can give green or red emission, which depends on the crystal field condition. Thus, tunable multicolor emission may be obtained by co-doping the above

86

J. Zhang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 165 (2016) 85–89

luminescent ions. In this paper, to explore new promising phosphors for LEDs and FEDs, a series of Ca3La6(SiO4)6:Tb3+, Mn2+ samples were synthesized, and their photoluminescence and cathodoluminescence (CL) properties were investigated. 2. Experimental Powder samples of Ca3 − yLa6 − x(SiO4)6:xTb3+, yMn2+ (CLS:xTb3+, yMn2+, 0 ≤ x ≤ 2, 0 ≤ y ≤ 0.45) were prepared by conventional solid-state reaction method. Stoichiometric amounts of the starting materials, CaCO3 (analytical reagent, AR), SiO2 (AR), La2O3 (4N), Tb4O7 (4N), and MnCO3 (AR), were thoroughly mixed and ground together by an agate mortar. The mixture was calcined at 1300 °C for 6 h in a reduction atmosphere (N2:H2 = 95:5). The phase purity was determined by using an ARL X'TRA powder Xray diffractometer (XRD) with Cu Kα radiation (λ = 1.5418 Å) operating at 40 kV and 35 mA. Diffuse reflection spectra (DRS) were obtained by a UV/visible spectrophotometer (UV-3600, SHIMADZU) using BaSO4 as a reference in the range of 200–700 nm. The morphology of the asprepared sample was inspected by field emission scanning electron microscope (FESEM, FEI, Quanta FEG). The luminescence spectra, decay curves, and external quantum efficiencies (QEs) were recorded on an FLS-920T fluorescence spectrophotometer with Xe 900 (450 W xenon arc lamp) as the light source. The CL properties of the samples were obtained using a modified Mp-Micro-S instrument. And the electrical characterizations for the phosphor were recorded with a Keithley 4200 and a Micromanipulator 6150 probe station. 3. Results and discussion The CLS compound belongs to hexagonal system with P63/m(176) space group and the crystal structure information has been given in Ref. [18]. Fig. 1 depicts the XRD patterns of the typical CLS:xTb3 +, yMn2+ (0 ≤ x ≤ 2, y = 0 and 0.21) phosphors. It can be observed that the hexagonal-structured CLS phase (JCPDS Card NO. 27-0078) has been formed, and the Tb3+/Mn2+ ions can be completely dissolved in the host lattices. However, a small amount of SiO2 phase is found for every XRD pattern, but this phase can't cause obvious effect on the luminescence spectra, so the following discussion on the luminescence will not involve this phase. Additionally, it can be seen from the XRD patterns that the diffraction peaks exhibit a continuous shift toward high angle direction with increasing Tb3+ concentration for the CLS:xTb3+ (0 ≤ x ≤ 2) samples. This is due to the Tb3+ of smaller ionic radius replaces La3+ with larger ionic radius, which also indicates the successful introduction of Tb3+ into the CLS host.

Fig. 1. XRD patterns of CLS:xTb3+, yMn2+ (0 ≤ x ≤ 2, y = 0 and 0.21).

Fig. 2 shows the SEM image of the typical CLS host. It can be found that the particles exhibit relatively dispersed morphology on the whole for the solid-state reaction preparation, but the particle scales are not uniform. To further observe the details of the particle surface, the inset of Fig. 2 shows the enlarged SEM image of one particle. It can be seen the particle surface is not smooth and some pits exist, this is owing to the agglomerate by lots of small particles to form a big particle in the calcination process. In addition, the average particle size is found to be around 10 μm, which is suitable for the practical application. Fig. 3 presents the diffuse reflection spectrum of the CLS from 200 to 700 nm. It shows a high reflectance in the visible range (400–700 nm), which is in agreement with the white daylight color of the CLS matrix. To determine the optical bandgap value of the CLS compound experimentally, the absorption spectrum of CLS (see the inset of Fig. 3) was obtained from its reflection spectrum using the Kubelka-Munk (K-M) function [19] F ðRÞ ¼ ð1−RÞ2 =2R ¼ K=S

ð1Þ

where R, K and S are the reflection, absorption and scattering coefficient, respectively. By extrapolating the K-M function to K/S = 0, the optical bandgap was determined to be about 3.9 eV. The excitation spectrum of the typical CLS:1.5Tb3 + sample monitored at 542 nm is represented in Fig. 4. The broad excitation band from 240 to 290 nm could be attributed to the f-d transition of Tb3 + [20]. The characteristic 4f-4f transitions of Tb3+ are observed between 300 and 500 nm, and the strongest excitation peak is located at 377 nm. From this excitation spectrum, it can be found the Tb3 +activated CLS phosphors can be efficiently excited by the NUV light, and the excitation peaks in the range of 340–390 nm could match with the NUV LED chip well. Fig. 5 shows the emission spectra of a series of CLS:xTb3 + (0.3 ≤ x ≤ 2) phosphors upon 377 nm excitation. The very weak emission peaks at 415 and 436 nm are respectively assigned to the 5D3-7F5 and 5D3-7F4 transitions of Tb3+ [21], but these peaks are almost quenched for high Tb3+ concentration due to the cross relaxation process as described by the formula Tb3(5D3) + Tb3(7F6) → Tb3(5D4) + Tb3(7F0) [21]. The 5D4-7FJ (J = 6 − 3) transitions of Tb3 + are located at 492, 542, 585, and 624 nm, respectively. With increasing Tb3+ content, the emission intensity of Tb3+ is increased gradually until x = 1.5, and beyond this Tb3+ concentration, the emission intensity begins to decrease due to the concentration quenching [22]. The external QE of CLS:1.5Tb3+ was measured to be about 20.2%. The Commission International del'Eclairage (CIE) chromaticity coordinates of the typical CLS:1.5Tb3+ were calculated from the emission spectra in the range from 400 to 650 nm to be (0.346, 0.582). Thus, a yellowish green emission is obtained as indicated by the CIE chromaticity diagram in Fig. 6 (Point 1) and

Fig. 2. SEM image of CLS.

J. Zhang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 165 (2016) 85–89

Fig. 3. Diffuse reflection spectrum of CLS, the inset shows its absorption spectrum of as calculated by the K-M formula.

the digital photograph under 365 nm ultraviolet lamp irradiation in the inset (a) of Fig. 6. For Tb3+-doped CLS, the thermal stability of luminescence was evaluated. Fig. 7(a) shows the emission spectra of the typical CLS:1.5Tb3+ sample upon 377 nm excitation obtained at different temperatures (from 25 to 200 °C). With increasing measure temperature, the emission intensity of Tb3+ demonstrates a gradual decrease, but the profiles of these emission spectra don't change. To observe this intensity decrease quantitatively, the corresponding temperature dependence of the Tb3+ integrated emission intensity is depicted in Fig. 7(b). It can be seen the emission intensity of Tb3+ changes little within the measure temperature of 100 °C, but exhibits a relatively fast decrease beyond this temperature. Nevertheless, the CLS:1.5Tb3+ sample shows a more excellent thermal stability compared with many Eu2 +/Ce3 + activated phosphors [20,23–25] since the brightness of the CLS:1.5Tb3+ sample is still above 90% of the initial value when the measure temperature reaches 200 °C. To investigate temperature quenching characteristics further, the activation energy (ΔE) was calculated using the Arrhenius equation [26]: IðT Þ ¼

I0

  ΔE 1 þ A  exp − kT

87

Fig. 5. Emission spectra of CLS:xTb3+ (0.3 ≤ x ≤ 2).

constant for a certain host, and k is the Boltzmann constant (8.629 × 10−5 eV). A plot of ln[(I0 / I) − 1] versus 1 / (kT) is depicted in Fig. 7(c). By linear fitting, the ΔE value was obtained to be 0.338 eV which is much larger than that of BaY2Si3O10:Tb3+ [20], indicating an excellent thermal stability. As we know, the ET process generally exists in Tb3+-Mn2+ codoped phosphors and abundant emitting light colors could be received by this way [27,28]. To evaluate the occurrence of ET between Tb3+ and Mn2+ in the CLS host, the inset (a) of Fig. 8 shows the excitation spectrum of CLS:0.1Mn2 + by monitoring 600 nm and the emission spectrum of CLS:1.5Tb3 + upon 377 nm excitation. The main excitation peaks at 341, 369, 405, 461, and 492 nm could be assigned to the transitions from the 6A1(6S) level to the 4E(4D), 4T2(4D), [4A1(4G), 4E(4G)], 4 T1(4G), and 4A2(4F) levels of Mn2+, respectively [29]. It is obvious the spectral overlap from 470 to 510 nm exists, so the ET from Tb3 + to Mn2+ may occur. Based on this point, a series of CLS:1.5Tb3+, yMn2+ (0 ≤ y ≤ 0.45) samples were prepared, and their emission spectra

ð2Þ

where I0 is the initial emission intensity, I(T) is the intensity at different temperatures, ΔE is activation energy of thermal quenching, A is a

3+

Fig. 4. Excitation spectrum of CLS:1.5Tb3+.

2+

Fig. 6. CIE chromaticity diagram for CLS:1.5Tb , yMn (Points 1–5 for y = 0, 0.09, 0.21, 0.33, and 0.45, respectively), insets (a–e) show the corresponding digital photograph of under 365 nm ultraviolet lamp irradiation.

88

J. Zhang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 165 (2016) 85–89

Fig. 9. Decay curves of CLS:1.5Tb3+, yMn2+ (0 ≤ y ≤ 0.45). 4

Fig. 7. (a) Emission spectra of CLS:1.5Tb3+ at different temperatures; (b) temperature dependence of the integrated emission intensity of CLS:1.5Tb3+; (c) activation plot for thermal quenching of CLS:1.5Tb3+.

T1(4G) level will increase, which will enhance the emission intensity of Mn2+. The above ET could be further verified by the decay lifetimes. Fig. 9 presents the decay curves of the typical CLS:1.5Tb3 +, yMn2 + (0 ≤ y ≤ 0.45) samples with 377 nm excitation and 542 nm emission. All the measured decay curves exhibit exponential feature and can be well-reproduced by a double-exponential function as I ¼ A1 expð−t=τ1 Þ þ A2 expð−t=τ2 Þ

ð3Þ

where τ1 and τ2 are the fast and slow components of the luminescent lifetimes, respectively, and A1 and A2 are their corresponding fitting parameters. The measured values are shown in Fig. 9. The average lifetimes could be determined by the following formula [30]

under 377 nm excitation are displayed in Fig. 8. The broad emission bands owing to the Mn2 + 4T1-6A1 transition are found from 500 to 700 nm, which overlap the 5D4-7FJ (J = 3 − 5) emissions of Tb3 +. With increasing Mn2+ concentration, the emission intensity of Tb3+ decays gradually and that of Mn2+ demonstrates a continuous increase. This observation indicates an efficient ET from Tb3+ to Mn2+. This ET process can be explained by the schematic of energy levels for Tb3 + and Mn2+ in the inset (b) of Fig. 8. When the Tb3+ ions are excited to the 5G6 state (377 nm excitation), the energy non-radiatively relaxes to 5D3, and further to 5D4 levels. The radiative transitions from 5D4 of Tb3+ to various 7FJ (J = 3 − 6) levels give rise to blue and green–red emissions. On the other hand, the transition energy between 5D4 and 7 F6 levels of Tb3 + is close to that between 6A1(6S) and 4T1(4G) levels of Mn2+, indicating the resonant non-radiative ET from Tb3+ to Mn2+ could take place. In this case, the electrons pumped to the Mn2 +

where τs and τs0 are the lifetimes of the sensitizer Tb3+ in the presence and absence of Mn2 +, respectively. From this result, it can be

Fig. 8. Emission spectra of CLS:1.5Tb3+, yMn2+ (0 ≤ y ≤ 0.45), inset (a) shows excitation spectrum of CLS:0.1Mn2+ and emission spectrum of CLS:1.5Tb3+, inset (b) shows schematic of ET process between Tb3+ and Mn2+.

Fig. 10. CL spectrum of CLS:1.5Tb3+, the inset shows its degradation property under constant electron beam bombardment (voltage = 5 kV, probe current = 50 mA).


bτN ¼ A1 τ21 þ A2 τ 22 =ðA1 τ 1 þ A2 τ2 Þ

ð4Þ

and the lifetime values were obtained to be 1.63, 1.16, 1.01, and 0.67 ms for y = 0.09, 0.21, and 0.45, respectively. Correspondingly, the ET efficiencies (ηT) were calculated respectively to be 29%, 38%, and 59% by the following equation [31] ηT ¼ 1−τ s =τ s0

ð5Þ

J. Zhang et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 165 (2016) 85–89

understood that the ET between Eu2+ and Mn2+ in CMGP is efficient. The external QE of the typical CLS:1.5Tb3+, 0.21Mn2+ was measured to be about 13.0% which was lower than that of CLS:1.5Tb3+, owing to the energy loss in the ET process. Correspondingly, the CIE chromaticity coordinates of CLS:1.5Tb3+, yMn2+ are obtained to be (0.483, 0.489), (0.499, 0.477), (0.506, 0.474), and (0.517, 0.468) for y = 0.09, 0.21, 0.33, and 0.45, respectively. The emitting light colors have been tuned to yellow from the yellowish green for y = 0–0.45 as can be seen in the CIE chromaticity diagram in Fig. 6 and the digital photographs under 365 nm ultraviolet lamp irradiation in the insets (b–e) of Fig. 6. For Tb3+-activated CLS phosphor, the CL properties are also investigated. The CL emission spectrum of the typical CLS:1.5Tb3 + sample under excitation of electron beam for 5 kV and 70 mA is given in Fig. 10. The 5D4-7FJ (J = 6 − 3) transition emissions of Tb3+ are found from 425 to 650 nm. Generally, the life of FEDs phosphor is related to its degradation property, so the degradation property of phosphor is an important factor for FEDs application. The inset of Fig. 10 presents the degradation property of CLS:1.5Tb3 + under constant electron beam bombardment with voltage of 5 kV and probe current of 50 mA. With increasing electron radiation time, the CL intensity decreases fast at first and changes little after 50 min. The brightness of the sample 90 min later becomes about 80% of the initial value. The degradation of the CL intensity mainly results from the effect of the continuous electron bombardment on the sample. This process will cause the accumulation of carbon at the surface, which can prevent low energy electron from reaching the phosphor grains and exacerbate surface charging [32]. 4. Conclusions In this paper, a series of CLS:Tb3+, Mn2+ phosphors were synthesized by conventional solid-state reaction method, and their spectral properties under NUV light and electron beam excitation were investigated. By monitoring 542 nm, the excitation spectrum of CLS:1.5Tb3+ exhibits strong excitation peaks in the region range of 340–390 nm, which can match well with the NUV LED chip. Upon 377 nm excitation, the f-f transition of Tb3+ are found from 400 to 650 nm and the 5D3-7FJ transitions are almost quenched for high Tb3+ concentration due to the cross relaxation process. Thus, a yellowish green emission is obtained in the CLS:1.5Tb3+. The investigation on the temperature dependence of the emission intensity for Tb3+ shows the Tb3+-activated CLS phosphor has a good thermal stability. By co-doping Mn2 + into CLS:Tb3 +, the emitting light color has been tuned to yellow by employing the ET from Tb3+ to Mn2+. These results indicate the CLS:Tb3+, Mn2+ phosphors could be promising candidate for LEDs applications. Moreover, the CL spectrum of CLS:Tb3+ reveals that the Tb3+-doped CLS can be also excited by electron beam, which implies it could apply to the FEDs.

89

Acknowledgments This work was supported by the Natural Science Foundation of Jiangsu Province of China (No. BK20140456) and Natural Science Foundation of the Jiangsu Higher Education Institutions of China (No. 14KJD140002).

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]

Z. Hou, G. Li, H. Lian, J. Lin, J. Mater. Chem. 22 (2012) 5254. X. Zhang, P. He, L. Zhou, J. Shi, M. Gong, Mater. Res. Bull. 60 (2014) 300–307. S. Ye, F. Xiao, Y. Pan, Y. Ma, Q. Zhang, Mater. Sci. Eng. R 71 (2010) 1–34. J. Sokolnicki, E. Zych, J. Lumin. 158 (2015) 65–69. G. Seeta Rama Raju, E. Pavitra, Phys. Chem. Chem. Phys. (2014). Z.Y. Zhao, Z.G. Yang, Y.R. Shi, C. Wang, B. Liu, G. Zhu, Y.H. Wang, J. Mater. Chem. C 1 (2013) 1407. Y. Shi, Y. Wang, Y. Wen, Z. Zhao, B. Liu, Z. Yang, Opt. Express 20 (2012) 21656–21664. Z. Xia, R.-S. Liu, K.-W. Huang, V. Drozd, J. Mater. Chem. 22 (2012) 15183–15189. W. Zhou, J. Han, X. Zhang, Z. Qiu, Q. Xie, H. Liang, S. Lian, J. Wang, Opt. Mater. 39 (2015) 173–177. S.K. Kim, K. Woo, K. Lim, H.J. Lee, Nanoscale 5 (2013) 9255–9263. W.-J. Yang, L. Luo, T.-M. Chen, N.-S. Wang, Chem. Mater. 17 (2005) 3883–3888. T. Jiang, X. Yu, X. Xu, H. Yu, D. Zhou, J. Qiu, Mater. Res. Bull. 51 (2014) 80. Z. Xia, R.-S. Liu, J. Phys. Chem. C 116 (2012) 15604. G. Wakefield, E. Holland, P.J. Dobson, J.L. Hutchison, Adv. Mater. 13 (2001) 1557–1560. P.H. Holloway, T.A. Trottier, B. Abrams, C. Kondoleon, S.L. Jones, J.S. Sebastian, W.J. Thomes, H. Swart, J. Vac. Sci. Technol. B 17 (1999) 758–764. M. Shang, G. Li, D. Yang, X. Kang, C. Zhang, J. Lin, J. Electrochem. Soc. 158 (2011) J125–J131. M. Xu, L. Wang, D. Jia, F. Le, J. Lumin. 158 (2015) 125–129. H.-C. Lin, C.-Y. Yang, S. Das, C.-H. Lu, J. Am. Ceram. Soc. 97 (2014) 1866–1872. Y.C. Wu, D.Y. Wang, T.M. Chen, C.S. Lee, K.J. Chen, H.C. Kuo, ACS Appl. Mater. Interfaces 3 (2011) 3195–3199. W.-R. Liu, C.C. Lin, Y.-C. Chiu, Y.-T. Yeh, S.-M. Jang, R.-S. Liu, B.-M. Cheng, Opt. Express 17 (2009) 18103–18109. J. Zhang, Y. Wang, Y. Wen, F. Zhang, B. Liu, J. Alloys Comp. 509 (2011) 4649–4652. J. Zhang, Y. Wang, Mater. Chem. Phys. 130 (2011) 1265–1269. W.-R. Liu, C.-H. Huang, C.-W. Yeh, J.-C. Tsai, Y.-C. Chiu, Y.-T. Yeh, R.-S. Liu, Inorg. Chem. 51 (2012) 9636–9641. J. Ruan, R.-J. Xie, S. Funahashi, Y. Tanaka, T. Takeda, T. Suehiro, N. Hirosaki, Y.Q. Li, J. Solid State Chem. 208 (2013) 50–57. Y. Wen, Y. Wang, B. Liu, F. Zhang, Y. Shi, Funct. Mater. Lett. 5 (2012) 1250048. R.J. Xie, N. Hirosaki, N. Kimura, K. Sakuma, M. Mitomo, Appl. Phys. Lett. 90 (2007) 191101. H. You, Y. Song, G. Jia, G. Hong, Opt. Mater. 31 (2008) 342–345. X. Xi, P. Dong, L. Han, R. Guan, G. Hou, J. Lumin. 159 (2015) 280–283. C.-H. Huang, T.-M. Chen, W.-R. Liu, Y.-C. Chiu, Y.-T. Yeh, S.-M. Jang, ACS Appl. Mater. Inter. 2 (2010) 259–264. T. Fujii, K. Kodaira, O. Kawauchi, N. Tanaka, H. Yamashita, M. Anpo, J. Phys. Chem. B 101 (1997) 10631–10637. P.I. Paulose, G. Jose, V. Thomas, N.V. Unnikrishnan, M.K.R. Warrier, J. Phys. Chem. Solids 64 (2003) 841–846. X.G. Xu, J. Chen, S.Z. Deng, N.S. Xu, J. Lin, J. Vac. Sci. Technol. B 28 (3) (2010) 490–494.