Upconversion luminescence of cerium-stabilized high temperature phase zirconia phosphors with a high Er³⁺ doping concentration

Abstract

Oxide upconversion (UC) materials usually have lower luminescence efficiency (LE) due to their low quenching concentration and high phonon energy. To solve this issue, we have investigated in detail cerium-stabilized high temperature phase zirconia based phosphors (Zr_(0.85−x)Ce_0.15O_(2−0.5x):xEr³⁺) with different doping concentrations of Er³⁺ and have achieved a red phosphor with an Er³⁺ concentration as high as 10 mol%, which will have broad prospects for its application as a new UC material. The results demonstrate that adjusting the host structure of the UC material can increase its quenching concentration and thereby the LE will be improved. The introduction of Ce⁴⁺ into ZrO₂ not only stabilizes its phase but also improves the LE of the ZrO₂ based UC material. The Er³⁺ coordination state changes with an increase of Er³⁺ concentration in the UC phosphors based on cerium-stabilized zirconia. The corresponding luminescence process translates into the cross relaxation from the excited state absorption, UC luminescence changes from green to red and the red emission intensity becomes stronger and stronger and achieves its maximum when the Er³⁺ concentration is 10 mol%.

---

Introduction

Rare-earth doped UC luminescence materials have attracted great scientific interest due to their potential applications in solid-state lasers,¹ optical fibre communication,² three-dimensional displays,^3,4 anti-counterfeit technology,⁵ bioimaging,^6,7 solar cells^8,9 and so on. UC materials not only demand high LE but also good physical and chemical properties for their practical application. Fluoride UC materials have attracted significant attention due to their lower phonon energy and higher LE,^3,6 but their poor chemical stability, mechanical strength and optical damage resistant properties have been the inevitable challenges for their actual application. Compared to fluoride, oxide based UC materials are more suitable for industrialization and practical application due to their pollution free fabrication process, simple manufacturing technique, excellent chemical stability and thermostability, but their lower LE blocks their further application. So, it has been a significant challenge to improve the LE of oxide UC materials.

Host and doped rare earth ions have important influences on the LE of UC materials under certain excitation sources. Generally speaking, lower phonon energies of substrates and higher rare earth doping concentrations will bring about higher LEs for UC materials, but high doping concentrations will cause serious fluorescence quenching due to the lower quenching concentrations for UC materials. Traditional UC polycrystalline or single crystal materials are often confined to a lower concentration (for example, the doping concentration of Er³⁺ in erbium doped UC materials is about 1 mol%),^3,10 and the lower quenching concentration has been the main barrier in improving the LE of many UC materials including oxide materials. To solve this bottleneck issue, taking cerium-stabilized zirconia UC phosphors as objects of study, which constitute cerium-stabilized zirconia doped with 15 mol% Ce⁴⁺ (marked with 15CSZ) used as the substrate and Er³⁺ as the active ion, several works on high rare-earth ion doping have been performed to improve the LE of oxide UC materials.

To achieve oxide UC materials doped with a high concentration of rare-earth ions, the substrate must have an adjusted structure and be easily able to form a good solid solution that has low phonon energy with other materials, however, few materials can satisfy this. In this paper, ZrO₂ with a low phonon energy is chosen as the base material for preparing these UC materials. ZrO₂ possesses a low phonon energy of about 470 cm⁻¹, good chemical and photochemical stability, a high melting point and so on, making it a good UC luminescence substrate.^5,11 As we all know, pure zirconia has three kinds of crystalline phase structure: monoclinic, tetragonal and cubic. The phase of zirconia changes with the preparation temperature and usage temperature, which causes the instability of the physical and chemical properties and hinders its application.^12–14 To obtain the stable ZrO₂ phase, stabilizers are added into ZrO₂, which can stabilize the crystalline phase and adjust the micro-structure of ZrO₂ at the same time. Luckily, a lot of research has been performed by scientists to investigate the phase transition mechanism of zirconia and to improve its stability. They have found that the high temperature phase of zirconia can be stabilized at room temperature by adding some stabilizers (Y₂O₃, CaO, Al₂O₃, MgO, CeO₂, etc.) to ZrO₂. Y₂O₃, as a stabilizer, is often used in the production of zirconia ceramics, however, CaO, MgO and CeO₂ are less widely used. Related studies have been reported by A. Feinber’s group,¹⁵ M. V. Swain’s group¹⁶ and P. Li’s group¹⁷ and so on. We chose CeO₂ as a stabilizer in the above oxides, which can form a miscible solid solution with ZrO₂. Not only can it stabilize the tetragonal phase of ZrO₂, but it also has low phonon energy (457 cm⁻¹).^17,18

15CSZ:xEr³⁺ phosphors are prepared via a reverse co-precipitation method and the influences of composition, structure and process conditions on the luminescence properties are investigated in detail. Their corresponding relations, phase change rules and the luminescence mechanisms of the phosphors are clarified. These will provide theoretical and technical support for raising the rare earth ion doping content and LE of UC materials and for developing new UC luminescent materials.

Experimental

Sample preparation

15CSZ:xEr³⁺ phosphors with different doping concentrations of Er³⁺ (x = 0.001, 0.005, 0.01, 0.05, 0.1, 0.15, 0.2) were prepared by a reverse co-precipitation method. The main materials of erbium oxide (Er₂O₃, 99.99%), cerium nitrate (Ce(NO₃)₃·6H₂O, AR), zirconium oxychloride (ZrOCl₂·8H₂O, AR) and nitric acid (HNO₃, AR) were purchased from Sinopharm Chemical Reagents Co. Ltd. Firstly, Er₂O₃ was weighed accurately according to the stoichiometric ratio, and then dissolved in concentrated HNO₃ until transparent. The excess acid was evaporated to obtain erbium nitrate. Then ZrOCl₂·8H₂O and Ce(NO₃)₃·6H₂O were weighed according to the stoichiometric ratio, and mixed with the previously prepared erbium nitrate, adding an appropriate volume of deionized water to prepare a cationic solution of about 0.5 mol L⁻¹, which was vigorously stirred until a homogeneous and transparent state was obtained. The mixture was slowly added to an excess of ammonium hydroxide NH₃·H₂O (2 mol L⁻¹), stirred until the titration was complete and the pH of the mixture was about 9–10, and then continuously stirred for 2 h to react completely. The precursor in the mixed solution was filtered and washed 2–3 times with deionized water. Afterwards, the precursor was dried at 105 °C for 12 h and ground in an agate mortar for 30 minutes. Then the precursor powder was moved to a crucible where it would be sintered at 1200 °C for 3 h in a muffle furnace. Finally, the samples were obtained.

Characterization

The powder X-ray diffraction (XRD) patterns were recorded by a Bruker/AXS D8-ADVANCE X-ray diffractometer using a copper target with Kα₁ radiation (λ = 0.15406 nm). The 2θ angle ranged from 10–90° and the working current and voltage were 40 mA and 40 kV, respectively. UC luminescence spectra were obtained with a computer-controlled TRIAX 320 spectrofluorimeter (Jobin-Yvon Corp.) under 980 nm LD (Coherent Corp.) excitation with a power of 100 mW and a monochromator with a Hamamatsu Photonics R955 photomultiplier tube connected to a PC. The Raman spectrum was recorded on a FT-Raman spectrophotometer (Invia Raman Microscope, RENISHAW) with an Ar ion 488.0 nm laser; the output power and focus diameter are 16 mW and 2 μm, respectively. All of the measurements were performed at room temperature.

Results and discussion

Influence of the preparation temperature on the phase and UC luminescence properties of phosphors

To investigate the influence of the preparation temperature on the phase and UC luminescence properties of phosphors, 15CSZ:0.1Er³⁺ phosphors were synthesized at 600 °C, 900 °C and 1200 °C for 3 h. Shown in Fig. 1 are XRD patterns of the phosphors synthesized at different temperatures. It can be seen that the phosphors have the same c-ZrO₂ solid solution crystalline phase, and Ce⁴⁺ and Er³⁺ completely replace Zr⁴⁺ and occupy the lattice sites of ZrO₂. Patra¹¹ et al. reported the preparation of ZrO₂:0.005Er³⁺ nanopowders with a mixture of monoclinic and tetragonal phases at 800 °C, 900 °C and 1000 °C. However, XRD patterns of the phosphors present no other phases except the pure c-ZrO₂ phase, which also confirms that the introduction of 15 mol% Ce⁴⁺ stabilizes the ZrO₂ phase.

Fig. 1 XRD patterns of the 15CSZ:0.1Er³⁺ phosphors prepared at different temperatures.

While the synthetic temperature increases, the intensity and shape of the XRD peaks are different, and the diffraction peaks become more and more high and sharp, indicating that the grown crystalline phase becomes more and more perfect. It is noteworthy that 15CSZ is a tetragonal zirconia phase, while 15CSZ:0.1Er³⁺ is a cubic zirconia phase. This is mainly due to the trivalent Er³⁺ ions in the zirconia substrate that can produce oxygen vacancies and have a coordination stabilizing effect on the zirconia substrate for this change. When the doping Er³⁺ concentration is lower the effect is not obvious, but when the doping Er³⁺ ion concentration is higher, Er³⁺ions produce more oxygen vacancies to make the tetragonal phase of zirconia transform into the cubic phase of zirconia.^12,19

Fig. 2 depicts the UC photoluminescence spectra of the 15CSZ:0.1Er³⁺ phosphors prepared at 600 °C, 900 °C and 1200 °C for 3 h under 980 nm excitation. With the sintering temperature increasing, the emission intensity of the red and green light become more and more strong. Especially, the emission intensity at 1200 °C is greatly enhanced compared to that at 600 °C and 900 °C (the green is enhanced by 113 and 89 times, the red is enhanced by 3600 and 353 times, respectively), which indicates that raising the preparation temperature could enhance the UC emission intensity. The reason for this may be (as mentioned above in the XRD analysis) that higher synthetic temperatures can bring better crystalline growth and less crystalline phase defects, which thereby bring lower rates of non-radiative relaxation so as to obviously improve the UC emission intensity.^20,21

Fig. 2 UC photoluminescence spectra of 15CSZ:0.1Er³⁺ phosphors prepared at different temperatures under 980 nm excitation.

XRD analysis of samples doped with different Er³⁺ content

Fig. 3 shows XRD patterns of the samples doped with different Er³⁺ content at 1200 °C for 3 h. ZrO₂:0.005Er³⁺ phosphors without Ce⁴⁺ have the m-ZrO₂ phase, which is consistent with what the related references reported.^10,13 According to the hard sphere theory, in terms of octahedral coordination structure, the cations and anions in a crystal can stably exist only when their radius ratio r⁺/r⁻ > 0.732. Conversely, the repulsive force is stronger than electrostatic attraction, which makes the octahedral coordination structure system not steady. For zirconia, due to r_Zr/r_O ≈ 0.564, the repulsive force between adjacent O–O bonds makes the structure not steady in the octahedral coordination structure. Therefore, zirconia can easily form the monoclinic phase structure with the coordination number being less than 8 under low temperature and the tetragonal and cubic phases with coordination numbers of 8 can only stably exist via lattice vibration under high temperature.²² However, the phase of Ce⁴⁺-stabilized 15CSZ:0.005Er³⁺ phosphors is a pure t-ZrO₂ solid solution phase, which demonstrates that 15 mol% Ce⁴⁺ stabilizes the high temperature phase of ZrO₂. Owing to the smaller ionic radius of Zr⁴⁺ (r = 0.084 nm) than that of Ce⁴⁺ (r = 0.097 nm), the introduction of Ce⁴⁺ enlarges the radius ratio of the cation/anion, causes lattice distortion and weakens the repulsive force between adjacent O–O bonds. This makes it possible for the tetragonal or cubic phases of the zirconia octahedral coordination structure to stably exist.

Fig. 3 (a) XRD patterns of 15CSZ:0.005Er³⁺ and ZrO₂:0.005Er³⁺ phosphors, (b) and (c) XRD patterns of 15CSZ:xEr³⁺ phosphors (x = 0.001, 0.005, 0.01, 0.05, 0.1, 0.15, 0.2).

Seen from Fig. 3b and c, it can be found that all 15CSZ:xEr³⁺ samples have no monoclinic phase. With the Er³⁺ doping concentration increasing, the phase of the phosphors changes from the tetragonal phase into the cubic phase, which can be explained as a certain synergy stabilizing effect played by Er³⁺ and Ce⁴⁺ in the sample phase.^12,19

UC photoluminescence spectra of samples doped with different Er³⁺ content

Fig. 4 shows the UC photoluminescence spectra of the 15CSZ:xEr³⁺ phosphors under 980 nm excitation. Two obvious characteristic emission bands can be seen containing the green emission ranging from 520–570 nm and the red emission ranging from 630–690 nm, which can be attributed to the ²H_11/2/⁴S_3/2 and ⁴F_9/2 → ⁴I_15/2 transition of Er³⁺, respectively.^10,23 To clearly observe the variation of the red and green intensity with the Er³⁺ doping content, Fig. 5 shows the variation of the UC emission peaks at 560 nm (the green) and 676 nm (the red) with the Er³⁺ doping content. With the Er³⁺ concentration increasing, the intensity of the green emission band ranging from 520–570 nm initially increases, then gradually decreases and achieves the highest value when the Er³⁺ doping content is 1 mol%. Meanwhile, the intensity of the red emission ranging from 630–690 nm initially also increases and then gradually decreases, achieving the highest value when the Er³⁺ concentration is 10 mol%. In all, the green emission is the main luminescence emission for samples with a low doping content, and by constantly increasing the Er³⁺ content, the luminescence changes from green to red, which also means a decrease of green emission and an increase of red emission. In fact, pure red UC emission is just what researchers pursue, for example, as red luminescence rather than green luminescence is more suitable for application in biological fluorescent labeling. Obviously, the emission intensity achieves the highest value while the Er³⁺ concentration is 10 mol%, which indicates that the Er³⁺ doping content breaks the low concentration limit in traditional UC materials and a high Er³⁺ doping content is realized. Realization of a high doping concentration can be explained by the following two possible reasons. On one hand, in the appropriate concentration scope, Er³⁺ substitutes the site of Zr⁴⁺ to form an electron-deficient centre (also called an oxygen vacancy) due to the absence of one positive charge in Er³⁺ compared with Zr⁴⁺. Each oxygen vacancy is surrounded by a pair of Er³⁺ ions, so it is highly difficult for the adjacent Er³⁺ ions to approach to form Er³⁺ clusters, which results in the obvious increase of the quenching concentration. On the other hand, it may be other defect species such as interstitial oxygen that compensate for the unbalanced charge due to the substitution of Ce³⁺, or Er³⁺ for Zr⁴⁺.

Fig. 4 UC photoluminescence spectra of 15CSZ:xEr³⁺ phosphors under 980 nm excitation.

Fig. 5 The variation of UC emission peaks at 560 nm (green) and 676 nm (red) with the Er³⁺ doping content.

Fig. 6 depicts the UC photoluminescence spectra of 15CSZ:0.005Er³⁺ and ZrO₂:0.005Er³⁺ phosphors under 980 nm excitation. The luminescence intensity of 15CSZ:0.005Er³⁺ is far higher than that of the ZrO₂:0.005Er³⁺ phosphors, and the green and red intensities improve 56 times and 5.8 times, respectively. So the introduction of 15 mol% Ce⁴⁺ not only stabilizes the zirconia phase but also greatly improves the UC luminescence intensity. One possible reason is that the introduction of CeO₂ reduces the phonon energy of substrate materials due to the low phonon energy (457 cm⁻¹) of CeO₂,²⁴ which can be explained via the Raman spectra of samples doped with Ce⁴⁺ ions or nothing in Fig. 7. The highest phonon energy of ZrO₂:0.005Er³⁺ is about 474 cm⁻¹, which is consistent with what the previous references reported,¹² however, the highest phonon energy of ZrO₂:0.005Er³⁺ doped with Ce⁴⁺ ions is about 458 cm⁻¹, indicating that the introduction of Ce⁴⁺ ions does lower the phonon energy of the substrate. Another possible reason is that the introduction of Ce⁴⁺ ions changes the phase of the substrate, and different phases have a great influence on the upconversion luminescence.¹³

Fig. 6 UC photoluminescence spectra of 15CSZ:0.005Er³⁺ and ZrO₂:0.005Er³⁺ phosphors under 980 nm excitation.

Fig. 7 Raman spectra of 15CSZ:0.005Er³⁺ and ZrO₂:0.005Er³⁺ phosphors.

Influences of phase on UC luminescence

To investigate the influences of the phase on UC luminescence, the emission spectra of different ZrO₂ phases are shown in Fig. 8. Generally speaking, there are two methods to get different ZrO₂ phases, one way is changing the doping concentration and another way is changing the calcination temperature. In our previous discussion, we have investigated the influence of temperature on the UC luminescence and a single phase was achieved under different temperatures due to the addition of a stabilizer. In this paper, we prepared ZrO₂ samples with different phases by changing the doping concentration. As clearly seen in Fig. 8, for the rare earth ion doped ZrO₂ phosphors with different phases, the intensity of the UC luminescence obviously changes, but the emission peak positions do not change. Therefore, it is obvious that different phases have great influences on the luminescence but the luminescence mechanism does not change.

Fig. 8 UC photoluminescence spectra of different ZrO₂ phases under 980 nm excitation.

UC luminescence mechanism analysis

Fig. 9 shows the energy level diagram of Er³⁺ and the possible UC luminescence mechanism. According to the previous reports, it is well known that the UC luminescence process mainly contains the ground state absorption (GSA), excited state absorption (ESA), cross relaxation (CR) and multi-phonon nonradiative relaxation and so on.^25–27

Fig. 9 The energy level diagram of Er³⁺ and the possible UC luminescence mechanism.

From Fig. 4 and 5 we can see that the leading luminescence is the green emission and the red emission is relatively weak. Moreover, the whole emission intensity is weak when the Er³⁺ doping content is lower in the samples. This can be explained by Er³⁺ ions absorbing fewer photons due to their low doping content, which decreases the number of Er³⁺ ions in the ²H_11/2/⁴S_3/2 and ⁴F_9/2 levels resulting in a weak emission intensity. For the green emission excited by 980 nm infrared light, Er³⁺ firstly absorbs a photon and transits to the metastable excited state ⁴I_11/2 level from the ground state, then transits immediately to the ⁴F_7/2 level after absorbing another photon. Er³⁺ ions located in the ⁴F_7/2 level quickly jump to the ²H_11/2 and ⁴S_3/2 levels via nonradiative relaxation and then most ions transit to the ground state resulting in a green emission (Fig. 9).¹³

However, red light can be obtained via two methods (shown in Fig. 9). The first way is that the Er³⁺ ions in the ⁴S_3/2 level transit to the ⁴F_9/2 level through nonradiative relaxation and then return back to the ground state with a red emission.²⁸ The second way is that Er³⁺ ions in the ⁴I_11/2 level transit to the ⁴I_13/2 level through nonradiative relaxation, then jump to the ⁴F_9/2 level after absorbing another photon and finally transit to the ground state with a red emission.

Owing to the low phonon energy, the ⁴S_3/2 and ⁴F_9/2 levels and the ⁴I_11/2 and ⁴I_13/2 levels both have a large energy level difference (about 3200 cm⁻¹ and 3600 cm⁻¹, respectively), therefore the chance of nonradiative relaxation for Er³⁺ ions is very low, which leads to a weak red emission.

As shown in Fig. 4 and 5, with the Er³⁺ doping concentration increasing, the green emission decreases and the red emission is enhanced, indicating that the number of Er³⁺ ions in the ²H_11/2/⁴S_3/2 level is decreasing and Er³⁺ ions in the ⁴F_9/2 level is gradually increasing. According to the previous discussion about the generation process of green and red emission, this change can happen only when the Er³⁺ ions in the ²H_11/2/⁴S_3/2 or ⁴F_7/2 levels transit to the ⁴F_9/2 level. However, this change can only happen between two Er³⁺ ions, which means that the luminescence mechanism of red emission has changed.

To clearly investigate whether the luminescence mechanism has changed, the red emission intensity of the 15CSZ:0.005Er³⁺ and 15CSZ:0.1Er³⁺ phosphors were tested under infrared light with different excitation powers and the experimental results are shown in Fig. 10. Theoretically, the relation between the UC emission intensity and excitation power can be described as the following:²⁹

I_em ∝ Pⁿ

I_em represents the UC emission intensity, P represents the excitation power and n represents the absorbed photon number originating from the excitation light source. The calculated line slope from log I_em and log P also represents the n value. In Fig. 10, the curve slope of the 15CSZ:0.005Er³⁺ and 15CSZ:0.1Er³⁺ phosphors at 676 nm is 1.41 and 1.71, respectively. This result indicates that the corresponding UC fluorescence processes of these two samples are both double photon absorption but have different formation mechanisms.¹²

Fig. 10 The logarithmic curves of the UC emission intensity of 15CSZ:0.005Er³⁺ and 15CSZ:0.1Er³⁺ phosphors at 676 nm and the corresponding excitation power.

So, what is the luminescence mechanism of the red light that leads to the increase of the red emission intensity and decrease of the green emission intensity? It is considered that Er³⁺ substitutes Zr⁴⁺ to form an oxygen vacancy, and each oxygen vacancy is surrounded by a pair of Er³⁺ ions with two kinds of coordination state. When the Er³⁺ doping content is low (0.1–1 mol%), these two Er³⁺ ions are in a coordination state where the cross relaxation (CR) cannot happen due to the large distance between them, however, when the Er³⁺ doping content is more than 1 mol%, the cross relaxation is able to happen due to the short distance between them,^19,30,31 which decreases the number of Er³⁺ ions in the ²H_11/2 and ⁴S_3/2 levels but increases that in ⁴F_9/2 and ⁴I_13/2. So the red emission is dominant in the luminescence and becomes stronger with an increase of the Er³⁺ doping concentration.

Conclusions

15CSZ:xEr³⁺ phosphors doped with different Er³⁺ content were prepared via a reverse co-precipitation method at 1200 °C for 3 h. A UC red phosphor with 10 mol% Er³⁺ concentration was achieved. This new UC material has a broad prospect for application and can be directly used or used as a kind of ceramic material at the same time. A new approach to improve the LE of UC materials is also achieved by increasing the quenching concentration through adjusting the host structure. The introduction of Ce⁴⁺ not only stabilizes the ZrO₂ phase but also improves the LE of ZrO₂ based UC materials. The red emission processes in UC luminescence involve double photon absorption, however, the luminescence mechanism varies with the Er³⁺ doping concentration. With the increase of the Er³⁺ doping content, the UC luminescence process translates from the excited state absorption into the cross relaxation effect, and the luminescence changes into a red emission from a green emission. The corresponding green and red emissions are the strongest when the Er³⁺ doping content is 1 mol% and 10 mol%, respectively. The realization of a high doping concentration for Er³⁺ may be attributed to oxygen vacancies and interstitial oxygen to compensate for the unbalanced charge in zirconia.

Acknowledgements

The authors acknowledge financial support from the Key Foundation of the Natural Science Foundation of Zhejiang Province (No. Y16E020043) and National Natural Science Foundation of China (Grant No. 51172165).

References

1. R. Scheps, Prog. Quantum Electron., 1996, 20, 271–358.
2. R. V. Perrella, D. P. D. Santos, G. Y. Poirier, M. S. Góes, S. J. L. Ribeiro, M. A. Schiavon and J. L. Ferrari, J. Lumin., 2014, 149, 333–340.
3. Y. Song, G. X. Liu, X. T. Dong, J. X. Wang, W. S. Yu and J. M. Li, RSC Adv., 2014, 4, 62802–62808.
4. J. S. Zhong, D. Q. Chen, W. G. Zhao, Y. Zhou, H. Yu, L. F. Chen and Z. G. Ji, J. Mater. Chem. C, 2015, 3, 4500–4510.
5. T. Pang, W. H. Cao, M. M. Xing, X. X. Luo and S. J. Xu, Chin. Sci. Bull., 2011, 56, 137–141.
6. G. Y. Chen, T. Y. Ohulchanskyy, R. Kumar, H. Agren and P. N. Prasad, ACS Nano, 2010, 4, 3163–3168.
7. D. Q. Chen, Y. Chen, H. W. Lu and Z. G. Ji, Inorg. Chem., 2014, 53, 8638–8645.
8. A. A. D. Adikaari, I. Etchart, P. H. Guéring, M. Bérard, S. R. P. Silva, A. K. Cheetham and R. J. Curry, J. Appl. Phys., 2012, 111, 094502.
9. A. S. Goncalves, M. S. Góes, F. Fabregat-Santiago, T. Moehl, M. R. Davolos, J. Bisquert, S. Yanagida, A. F. Nogueira and P. R. Bueno, Electrochim. Acta, 2011, 56, 6503–6509.
10. L. A. Gómez, G. S. Maciel, C. B. D. Araújo and A. Patra, J. Appl. Phys., 2008, 103, 053507.
11. R. Reisfeld, M. Zelner and A. Patra, J. Alloys Compd., 2000, 300, 147–151.
12. D. Solis, E. de la Rosa, O. Meza, L. A. Diaz-Torres, P. Salas and C. Angeles-Chavez, J. Appl. Phys., 2010, 108, 023103.
13. A. Patra, C. S. Friend, R. Kapoor and P. N. Prasad, Appl. Phys. Lett., 2003, 83, 284–286.
14. Y. Q. Zhou, H. Wang, B. Xu and L. L. Cao, J. Lumin., 2015, 161, 288–292.
15. A. Feinber and C. H. Perry, J. Phys. Chem. Solids, 1981, 42, 513–518.
16. M. V. Swain, R. C. Garvie and R. H. J. Hannink, J. Am. Ceram. Soc., 2006, 66, 358–362.
17. P. Li, I. W. Chen and J. E. Penner-Hahn, J. Am. Ceram. Soc., 1994, 77, 1281–1288.
18. C. Bozo, F. Gaillard and N. Guilhaume, Appl. Catal., A, 2001, 220, 69–77.
19. A. Patra, C. S. Friend, R. Kapoor and P. N. Prasad, J. Phys. Chem. B, 2002, 106, 1909–1912.
20. X. Wang, X. G. Kong, G. Y. Shan, Y. Yu, Y. J. Sun, L. Y. Feng, K. F. Chao, S. Z. Lu and Y. J. Li, J. Phys. Chem. B, 2004, 108, 18408–18413.
21. X. Wang, X. G. Kong, Y. Yu, Y. J. Sun and H. Zhang, J. Phys. Chem. B, 2007, 111, 15119–15124.
22. A. Bogicevic, C. Wolverton, G. M. Crosbie and E. Stechel, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 014106.
23. G. Y. Chen, Y. G. Zhang, G. Somesfalean, Z. G. Zhang, Q. Sun and F. P. Wang, Appl. Phys. Lett., 2006, 89, 3105.
24. H. Guo, J. Solid State Chem., 2007, 180, 127–131.
25. G. Y. Chen, T. Y. Ohulchanskyy, A. Kachynski, H. Agren and P. N. Prasad, ACS Nano, 2011, 5, 4981–4986.
26. H. J. Liang, Y. D. Zheng, G. Y. Chen, L. Wu, Z. G. Zhang and W. W. Cao, J. Alloys Compd., 2011, 509, 409–413.
27. D. Q. Chen, L. Liu, P. Huang, M. Y. Ding, J. S. Zhong and Z. G. Ji, J. Phys. Chem. Lett., 2015, 6, 2833–2840.
28. A. C. Yanes, A. Santana-Alonso, J. Méndez-Ramos, J. del-Castillo and V. D. Rodríguez, J. Alloys Compd., 2009, 480, 706–710.
29. H. M. Noh, H. K. Yang, B. K. Moon, B. C. Choi, J. H. Jeong and H. Choi, J. Nanosci. Nanotechnol., 2013, 13, 4006–4009.
30. F. Vetrone, J. C. Boyer, J. A. Capobianco, A. Speghini and M. Bettinelli, J. Phys. Chem. B, 2003, 107(5), 1107–1112.
31. L. Sun, C. H. Yang, A. H. Li and Y. H. Xu, J. Appl. Phys., 2009, 105, 043512.

---