Transparent Na₅Gd₉F₃₂:Er³⁺ glass-ceramics: enhanced up-conversion luminescence and applications in optical temperature sensors

Abstract

New rare earth ion-doped nanocomposite materials, Na₅Gd₉F₃₂:Er³⁺ glass-ceramics, were fabricated via a traditional melt-quenching technique and subsequent heat treatments. Their microstructural and optical properties were systemically investigated by XRD, TEM, HRTEM techniques, absorption, up-conversion spectra and luminescence lifetime measurements. Excited by 980 nm laser, both glass and glass-ceramic samples presented characteristic red and green up-conversion emissions of Er³⁺ ions. After crystallization, Er³⁺ ions were incorporated into the precipitated Na₅Gd₉F₃₂ nanocrystals. Compared with precursor glasses, the red up-conversion intensity of glass-ceramics was enhanced by 1300 times, and the luminescence lifetime was also prolonged. The fluorescence intensity ratio technique was utilized to carry out the optical thermometry based on the green up-conversion luminescence behavior of Er³⁺ ions. Results manifest that Na₅Gd₉F₃₂:Er³⁺ glass-ceramics would have potential applications in optical temperature sensors with high sensitivity.

---

Introduction

Fluoride nanocrystal-based oxyfluoride glass-ceramics, a new type of nanocomposite material,^1–18 have received much interest in recent years. Generally, glass-ceramics are fabricated by a traditional melt-quenching technique with subsequent crystallization processes. They are isotropic and possess the ability to be shaped into diverse geometrical structures. By proper heat treatment, one or more crystalline phases can be participated through in situ growth in a glass matrix.^19–21 Thus, the advantages of fluoride nanocrystals can be achieved with the merit of glass preserved. That is to say that glass-ceramics combine the high chemical, mechanical and thermal stability of glasses with the low phonon energy and strong crystal field environment of fluoride nanocrystals.^3–17

Inspiringly, doped rare earth (RE) ions will preferentially get incorporated into the as-synthesized fluoride nanocrystals phase^8–17 and exhibit excellent luminescent behavior, which makes RE ion-doped fluoride glass-ceramics potential candidates for optoelectronic devices, such as in lighting,²² multi-color displays,¹¹ sensors¹⁵ and scintillators.²³ In this case, numerous RE fluoride glass-ceramics have been fabricated and explored, such as La, Gd, Yb, Lu and Y-based fluoride glass-ceramics.^{13–17,22–26}

The luminescent properties of RE ion-doped fluoride glass-ceramics remarkably depend on the crystalline phase precipitated in glass matrix. The sodium–gadolinium–fluoride compounds (Na_xGd_yF_x+3y), particularly NaGdF₄ and Na₅Gd₉F₃₂, are identified as efficient up-conversion hosts for RE ions doping. The fabrication and luminescent properties of RE ion-doped NaGdF₄ and Na₅Gd₉F₃₂ nanoparticles have frequently attracted attention.^27–30 However, the application of such fluoride nanocrystals is hampered by their insufficient thermal stability leading to phase transition or decomposition during the reheating process.^28–30 Fortunately, this disadvantage may be compensated by integrating Na_xGd_yF_x+3y nanocrystals in a glassy matrix considering the good properties of glass-ceramics mentioned above. Numerous investigations on NaGdF₄ glass-ceramics have been reported,^3,31 particularly on the up-conversion luminescence behavior of Er³⁺-doped ones.³ However, only one report on Na₅Gd₉F₃₂ glass-ceramics was published.³² Therefore, it is promising to investigate whether Na₅Gd₉F₃₂ glass-ceramics are thermal stable Na_xGd_yF_x+3y compound for RE up-conversion luminescence.

Er³⁺ ions are the most excellent up-conversion luminescent centers³³ exhibiting visible up-conversion luminescence pumped by near-infrared (NIR) laser. They can be used as probes for optical detection of magnetic field.³⁴ In addition, Er³⁺ ions have a couple of thermally coupled energy levels (TCEL), ²H_11/2 and ⁴S_3/2 levels. As a result, there has been growing interest in the optical thermometry of glass-ceramics doped with Er³⁺ ions,^15,16 since these Er³⁺-doped glass-ceramics can be used as optical temperature sensors. Moreover, no reports on optical thermometry of Er³⁺-doped Na₅Gd₉F₃₂ glass-ceramics have been published. Therefore, it is incumbent upon us to research the structural, up-conversion luminescent properties and temperature sensing properties of Er³⁺-doped Na₅Gd₉F₃₂ glass-ceramics.

In this study, Er³⁺-doped novel transparent oxyfluoride glass-ceramics containing Na₅Gd₉F₃₂ nanocrystals were fabricated and the unique up-conversion properties of Er³⁺ ions were investigated in detail. No phase transition occurs during different heat treatment processes, which suggests that Na₅Gd₉F₃₂ glass-ceramics have better thermal stability than the reported Na₅Gd₉F₃₂ nanocrystals.²⁹ Enhanced up-conversion emissions and prolonged luminescence lifetimes of Er³⁺ ions indicate that Er³⁺ ions have been preferentially incorporated into Na₅Gd₉F₃₂ nanocrystals. The optical thermometry results of temperature-dependent green up-conversion spectra suggest that the Na₅Gd₉F₃₂ glass-ceramics with high sensitivity can be good candidates for optical temperature sensors.

Experimental

Glass samples with nominal composition 45SiO₂–10Na₂CO₃–15Al₂O₃–4CaCO₃–18NaF–8GdF₃–0.2ErF₃ (in mol%) were prepared by the melt-quenching method. SiO₂, Al₂O₃, Na₂CO₃, NaF (A.R., all from Sinopharm Chemical Reagent Co., Ltd.), and high-purity GdF₃ and ErF₃ (99.99%, from AnSheng Inorganic Materials Co., Ltd.) were used as starting materials. The well ground stoichiometric chemicals were put into a covered alumina crucible and melted at 1500 °C for 1 h in air atmosphere. The melt was poured onto a 300 °C preheated stainless-steel plate and then pressed by another plate to get solid samples. After annealing at 450 °C for 5 h to release internal stresses, precursor glasses (labelled as PG) were formed. Subsequently, PG were subjected to heat treatment for 2 h at 590 °C, 610 °C and 630 °C to fabricate transparent glass-ceramics, which were labelled as GC590, GC610 and GC630, respectively. All samples were polished optically with a thickness of 2 mm for further characterization.

X-ray diffraction (XRD) patterns were obtained on the Philips X' Pert PRO SUPER X-ray diffraction apparatus (40 kV, 40 mA) with Cu Kα radiation (λ = 0.154056 nm) over the angular range of 10° ≤ 2θ ≤ 80° in a step size of 0.0167°. For XRD measurement, the lump glasses were ground to a fine powder in an agate mortar. Transmittance spectra were measured on the U-3900 Ultraviolet-Visible (UV-VIS) spectrophotometer. The microstructure of glass-ceramics was analyzed using the JEM-2100F transmission electron microscope (TEM) operated at an accelerating voltage of 200 kV (JEOL Ltd.). Up-conversion spectra were measured on the FS920 spectrofluorometer (Edinburgh Instruments) with a 980 nm laser (300 mW mm⁻²) as the excitation source. The spot size of the 980 nm laser was about 1 × 6 mm². In addition, the pump power was adjusted through neutral density filters. The temperature of the sample fixed on a copper post was controlled over the range of 300–510 K by a temperature controller (FOTEK MT48-V-E, Taiwan) with a type-K thermocouple and a heating tube. Decay curve measurements were performed on the SBP500 monochromator (Zolix Instruments) coupled to the Hamamatsu R928 photomultiplier with the Tektronix TDS5052 oscilloscope.

Results and discussion

Fig. 1(a) shows the XRD patterns of PG, GC590, GC610 and GC630 samples. PG does not exhibit any discrete diffraction peaks, confirming its amorphous nature. For GC590, GC610 and GC630 samples, their diffraction peaks match well with those of the cubic Na₅Gd₉F₃₂ (JCPDS card no. 27-0698), indicating that cubic Na₅Gd₉F₃₂-based glass-ceramics were successfully fabricated. The GC630 sample crystallized at 630 °C still maintains cubic Na₅Gd₉F₃₂ phase, indicating that Na₅Gd₉F₃₂ glass-ceramics have better thermal stability than nano-sized cubic Na₅Gd₉F₃₂ spheres.²⁹ The size of Na₅Gd₉F₃₂ nanocrystals in glass-ceramics can be calculated by the following Scherrer equation:³⁵

D = kλ/β cosθ (1)

where k = 0.89, λ = 0.154056 nm, which represents the wavelength of Cu Kα radiation, θ is the Bragg angle and β represents the corrected half width of diffraction peak. The mean crystalline sizes estimated are about 14, 17 and 24 nm for Na₅Gd₉F₃₂ nanocrystals in GC590, GC610 and GC630 (listed in Table 1), respectively.

Fig. 1 (a) XRD patterns of PG, GC590, GC610 and GC630 samples, and the reference data of JCPDS card no. 27-0698 for cubic Na₅Gd₉F₃₂. (b) Transmission spectra of PG, GC590, GC610 and GC630 samples. (c) TEM and (d) HRTEM images of GC610; the inset is the SAED pattern. Photographs of (e) PG, (f) GC590, (g) GC610 and (h) GC630 samples.

Table 1 Synthesized and heat treatment temperature, mean crystalline sizes, up-conversion luminescence and lifetimes of all samples

	Synthesized temperature	Heat treatment temperature	Mean crystalline sizes (nm)	Transmittance at 600 nm	Enhanced factor of red emission (vs. PG)	Lifetime of ^4F9/2 → ^4I15/2 transition (ms)
PG	1500 °C, 1 h	—	—	90%	—	0.51
GC590	1500 °C, 1 h	590 °C, 2 h	14	90%	130 times	2.81
GC610	1500 °C, 1 h	610 °C, 2 h	17	74%	570 times	3.58
GC630	1500 °C, 1 h	630 °C, 2 h	24	34%	1300 times	3.80

---

The transmission spectra of PG, GC590, GC610 and GC630 samples are shown in Fig. 1(b) and the transparence of GC610 sample is up to 74% at 600 nm. The absorption peaks located at 379, 488, 525 and 652 nm are ascribed to the transitions from ground ⁴I_15/2 state to ⁴G_11/2, ⁴F_7/2, ²H_11/2 and ⁴F_9/2 excited states of Er³⁺, respectively.¹⁴ Photographs and transmittances of PG and all GC samples are given in Fig. 1(e)–(h) and Table 1, respectively. It can be seen that Na₅Gd₉F₃₂ glass-ceramics are transparent with excellent uniformity, indicative of homogeneous crystallization of these samples. According to Rayleigh–Gans theory,⁵ except for the size of nanocrystals, the difference in refractive index between the crystal phase and the glass phase is also an important parameter for the transmittance of samples. For GC590 and GC610 samples, as the size of the precipitated crystals is much smaller than the wavelength of visible and near-infrared light, glass-ceramics still maintain a good transparency in visible to near-infrared region (Fig. 1(b), (f) and (g)). However, with the increase in heat-treatment temperature, the size of nanocrystals increases to some extent, and the influence of refractive index between the crystal phase and the glass phase plays a more significant role in transmittance from glass-ceramics. Therefore, the transparence of GC630 sample decreased (Fig. 1(b) and (h)) due to the strong light scattering of Na₅Gd₉F₃₂ nanocrystals and the large difference in the refractive index between Na₅Gd₉F₃₂ crystal phase and glass phase.

TEM and high-resolution TEM (HRTEM) images of GC610 sample are shown in Fig. 1(c) and (d), respectively. TEM bright-field micrograph reveals that Na₅Gd₉F₃₂ nanocrystals are homogeneously dispersed in amorphous glassy phase. The corresponding selected area electron diffraction (SAED) patterns of GC610 sample indicate that glass-ceramics are composite materials that contain amorphous glasses and poly-nanocrystals. The crystal size of Na₅Gd₉F₃₂ nanocrystals in GC610 sample is about 17 nm, in accordance with that estimated by the Scherrer equation. The HRTEM image clearly displays the resolved lattice fringes and the value of the associated interplanar spacing d is about 0.333 nm, which corresponds to (111) crystal plane of cubic Na₅Gd₉F₃₂ (d₍₁₁₁₎ = 0.321 nm).

The up-conversion spectra (λ_ex = 980 nm, 300 mW mm⁻²) of PG, GC590, GC610 and GC630 samples are given in Fig. 2(a). Green emission bands (located at 525 and 540 nm) and red emission bands (located at 660 nm) are assigned to ²H_11/2/⁴S_3/2 → ⁴I_15/2 and ⁴F_9/2 → ⁴I_15/2 transitions of Er³⁺, respectively.³⁵ Distinct Stark splitting at 540 and 660 nm in GC samples are observed after crystallization. Moreover, the up-conversion emissions in GC samples are enhanced prominently. Compared with PG, green emissions in GC590, GC610 and GC630 samples enhance by 110, 470 and 1000 times, respectively, whereas red emissions in GC590, GC610 and GC630 samples enhance by 130, 570 and 1300 times (listed in Table 1), respectively. The evident Stark splitting and enhanced up-conversion luminescence may originate due to the incorporation of Er³⁺ ions into cubic Na₅Gd₉F₃₂ nanocrystals with variant surroundings of Er³⁺ ions after crystallization, such as changes in symmetry and phonon energy.^5,14

Fig. 2 (a) Up-conversion spectra of PG, GC590, GC610 and GC630 samples; dependence of up-conversion on pump intensity for (b) PG and (c) GC610 samples.

For the up-conversion process, intensity of up-conversion emission I is proportional to the nth power of pump power density P:³⁵

I ∝ Pⁿ (2)

where n is the number of pump photons absorbed per up-converted photon emitted. A plot of log I versus log P yields a straight line with slope n. The power dependences of up-conversion of PG and GC610 samples are shown in Fig. 2(b) and (c), respectively. The obtained n values indicate that both red and green up-conversion emissions are caused by two-photon processes.

The excited states for up-conversion can be populated by two famous mechanisms: (1) excited state absorption (ESA) and (2) energy transfer (ET).³⁵ The ESA process is a single-ion process. Therefore, the rise time of the decay curve of luminescence pumped by the ESA process will be equal to zero. In contrast, the ET process involves two ions, and the ET rate will increase with short Er³⁺–Er³⁺ distances. Hence, the rise time of the decay curve of luminescence pumped by the ET process will be at the same level of lifetime of related metastable levels.

It is well known that glass-ceramics with preferential enrichment of RE ions in the precipitated fluoride nanocrystals can exhibit excellent luminescent behavior and possess a longer lifetime.¹⁴ To further prove the preferential incorporation of Er³⁺ ions into cubic Na₅Gd₉F₃₂ nanocrystals and to understand the up-conversion mechanisms more clearly, the decay curves of up-conversion emissions of PG and GC samples were investigated.

Fig. 3(a) shows the decay curves of luminescence at 660 nm (⁴F_9/2 → ⁴I_15/2) emissions of Er³⁺ in PG and GC samples. The lifetimes are characterized by average lifetime ([small tau, Greek, macron]) derived by¹⁶

where I(t) is the emission intensity at time t. The average lifetimes of ⁴F_9/2 level are about 0.51, 2.81, 3.58 and 3.80 ms for PG, GC590, GC610 and GC630 samples (listed in Table 1), respectively.

Fig. 3 Decay curves for (a) ⁴F_9/2 → ⁴I_15/2 (660 nm) emission of Er³⁺ in PG, GC590, GC610 and GC630 samples (λ_ex = 980 nm). (b) The partial enlarged image of (a). (c) Energy level diagram of Er³⁺ and possible up-conversion mechanisms.

Prolonged lifetimes for up-conversion emissions in GC samples indicate the reduced non-radiative relaxation rate of Er³⁺ in GC samples and prove the preferential enrichment of Er³⁺ ions into Na₅Gd₉F₃₂ nanocrystals with lower phonon energy. In addition, the rise times at the beginning of decay curves of GC samples are different from that of PG. As shown in Fig. 3(b), the partial magnification of Fig. 3(a), the rise time of PG is almost zero. For all GC samples, the rise times are approximately 0.3 ms for 660 nm emission. Such phenomenon proves that ESA processes are responsible for up-conversion in PG, whereas ET processes are responsible for up-conversion in GC.

In summary, the evident Stark splitting, enhanced characteristic up-conversion emissions and prolonged lifetimes in GC samples demonstrate the preferential incorporation of Er³⁺ into Na₅Gd₉F₃₂ nanocrystals. In other words, for PG sample, Er³⁺ ions are homogeneously dispersed in glassy phase with long Er³⁺–Er³⁺ distances. For GC samples, most Er³⁺ ions are incorporated into Na₅Gd₉F₃₂ nanocrystals with short Er³⁺–Er³⁺ distances. The changes in up-conversion spectra are caused by the different surroundings of Er³⁺ ions and the different up-conversion approaches in PG and GC.

On the basis of energy level diagram of Er³⁺ and above discussions, the up-conversion mechanisms accounting for green and red emissions are analyzed and illustrated in Fig. 3(c). Er³⁺ ion is first excited through ground state absorption (Fig. 3(c): GSA) process from ⁴I_15/2 level to ⁴I_11/2 level by absorbing one 980 nm photon. Subsequently, the ion at ⁴I_11/2 level is populated to luminescent levels by different approaches in PG and GC.

For PG sample, ESA processes are dominant because of the longer Er³⁺–Er³⁺ distances of the homogeneous dispersed Er³⁺ ions in glassy phase. Er³⁺ ion at ⁴I_11/2 level absorbs another excitation 980 nm photon and is pumped to ⁴F_7/2 level (Fig. 3(c): ESA1).³⁵ Due to the small energy gap between ⁴F_7/2 level and ²H_11/2, ⁴S_3/2 levels, Er³⁺ ion at ⁴F_7/2 level can easily and non-radiatively relax to the relatively stable ²H_11/2 and ⁴S_3/2 levels and then produce green up-conversion emissions. The red up-conversion emitting ⁴F_9/2 level can be pumped by the following two processes: one is the non-radiative relaxation from the excited ⁴S_3/2 state^11,35 and the other is that Er³⁺ absorbs a second 980 nm photon from ⁴I_13/2 level (Fig. 3(c): ESA2).¹¹

However, for GC samples, ET processes are primarily responsible for up-conversion emission attributing to the fact that Er³⁺ ions incorporate into Na₅Gd₉F₃₂ nanocrystals preferentially with shorter Er³⁺–Er³⁺ distances, lower phonon energy environment and higher crystallization degree. These advantages will reduce the multi-phonon non-radiative relaxation rate and result in enhanced up-conversion luminescence (1300 times).⁶ Green luminescent levels can be pumped as follows: an excited Er³⁺ ion at ⁴I_11/2 level relaxes to ⁴I_15/2 level non-radiatively and transfers its excitation energy to a neighboring Er³⁺ ion at the same level, promoting the latter to ⁴F_7/2 level: ⁴I_11/2 + ⁴I_11/2 → ⁴I_15/2 + ⁴F_7/2 (Fig. 3(c): ET1).³⁵ Red luminescent ⁴F_9/2 level can be populated by the well-known ET2 process: ⁴I_11/2 + ⁴I_13/2 → ⁴I_15/2 + ⁴F_9/2.³⁵

The enhanced up-conversion luminescence and good thermal stability (mentioned in XRD analyses) make Na₅Gd₉F₃₂ glass-ceramics promising for highly sensitive optical thermometry. As for optical thermometry, there are three main optical techniques: fluorescence lifetime, amplified spontaneous emissions and fluorescence intensity ratio (FIR) techniques.^15,36–38 Compared with the former two, FIR technique has improved measurement accuracy and widened operating temperature range. It is built on the temperature dependence of the FIR value of transitions from TCEL. Detailed FIR technique can be found in our previous study.¹⁶ FIR technique can reduce the dependence on measurement conditions and has been considered as a promising approach for temperature sensing.

In this case, the temperature-dependent green up-conversion spectra of Er³⁺-doped Na₅Gd₉F₃₂ glass-ceramics were measured by the FIR value of the TCEL (²H_11/2 and ⁴S_3/2).³⁹ Fig. 4(a) depicts the normalized green up-conversion spectra (normalized at 522 nm) of GC630 sample measured at different temperatures from 300 to 510 K excited by 980 nm laser with a pump power density of 20.6 mW mm⁻². It is clear that the relative intensity ratio of ²H_11/2 → ⁴I_15/2 transition versus ⁴S_3/2 → ⁴I_15/2 transition increases monotonously with the elevation in temperature. This is because the population of the ²H_11/2 and ⁴S_3/2 levels of Er³⁺ ion are thermally coupled, and their populations are in accordance with the Boltzmann distribution law. Moreover, the overlap between these two emission bands is small and can be ignored, which favors the measurement accuracy of emission intensities. Both of these characteristics make Na₅Gd₉F₃₂:Er³⁺ glass-ceramics potential probes for temperature sensing.

Fig. 4 (a) Green up-conversion spectra of GC630 normalized at 522 nm under the excitation of a 980 nm diode laser at different temperatures from 300 to 510 K. (b) Monolog plot of the FIR as a function of inverse absolute temperature. (c) Temperature dependence of FIR between ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 transitions of Er³⁺. (d) The relative sensitivity S_R and the absolute sensitivity S_A of GC630 under the excitation of a 980 nm diode laser at various temperatures from 300 to 510 K.

Fig. 4(b) presents the monolog plot of the FIR value of emissions from ²H_11/2 and ⁴S_3/2 as a function of inverse absolute temperature in the range of 300–510 K. The experimental data are fitted to a straight line with the slope of about −998.6. Fig. 4(c) illustrates the temperature dependence of the FIR value in the range of 300–510 K for GC630 sample, and the curve can be well fitted with the following equation:^16,25

FIR = B exp(−ΔE/K_BT) = 3.30 exp(−998.6/T) (4)

where B is the pre-exponential parameter; ΔE is the effective energy gap between ²H_11/2 and ⁴S_3/2; K_B is the Boltzmann constant and T is the absolute temperature. According to the fitting results, the effective energy gap ΔE between ²H_11/2 and ⁴S_3/2 levels of Er³⁺ ion can be obtained with a value of 695 cm⁻¹. It is known that the relative sensitivity S_R and the absolute sensitivity S_A are two important parameters to evaluate the property of the sensors for temperature sensing.¹⁶ Thus, the S_R and S_A of GC630 sample are investigated and their fitting results as a function of temperature from 300 to 510 K are described in Fig. 4(d). The calculated S_R value is 998.6/T²% K⁻¹. In addition, the S_A value increases with the temperature and is up to the maximum of 17.9 × 10⁻⁴ K⁻¹ at 499.3 K. Such high sensitivity indicates that Na₅Gd₉F₃₂:Er³⁺ glass-ceramics may be used as optical temperature sensors.

Conclusions

Er³⁺-doped transparent Na₅Gd₉F₃₂-based glass-ceramics were successfully fabricated by the melt-quenching method with subsequent crystallization processes. Phase transition did not occur with the increasing heat treatment temperature, which certified the good thermal stability of Na₅Gd₉F₃₂ glass-ceramics. The evident Stark splitting, enhanced characteristic up-conversion emissions (1300 times for red emission, 1000 times for green emission), slower rise time and prolonged luminescence lifetime in glass-ceramics reveal the preferential incorporation of Er³⁺ into Na₅Gd₉F₃₂ nanocrystals with low phonon energy and shorter Er³⁺–Er³⁺ distances after crystallization. ESA and ET processes are dominant up-conversion mechanisms in PG sample and GC samples, respectively. The analyses of temperature-dependent green up-conversion manifest the thermal-stable Na₅Gd₉F₃₂ glass-ceramics, as novel up-conversion functional materials, may have prospects for optical temperature sensors in the future.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (Grant No. 11374269).

Notes and references

1. C. G. Lin, C. Bocker and C. Rüssel, Nano Lett., 2015, 15, 6764–6769.
2. S. Liu, D. Q. Chen, Z. Y. Wan, Y. Zhou, P. Huang and Z. G. Ji, RSC Adv., 2016, 6, 71176–71187.
3. F. X. Xin, S. L. Zhao, L. H. Huang, D. G. Deng, G. H. Jia, H. P. Wang and S. Q. Xu, Mater. Lett., 2012, 78, 75–77.
4. Z. Y. Zhao, C. Liu, M. L. Xia, Q. Y. Yin, X. J. Zhao and J. J. Han, J. Alloys Compd., 2017, 701, 392–398.
5. X. S. Qiao, X. P. Fan, J. Wang and M. Q. Wang, J. Non-Cryst. Solids, 2005, 351, 357–363.
6. W. Xu, X. Y. Gao, L. J. Zheng, Z. G. Zhang and W. W. Cao, Opt. Express, 2012, 20, 18127–18137.
7. Y. P. Tai, H. Y. Wang, H. Wang and J. T. Bai, RSC Adv., 2016, 6, 4085–4089.
8. D. Q. Chen, Y. S. Wang, Y. Yu and P. Huang, Appl. Phys. Lett., 2007, 91, 051920.
9. D. Q. Chen, Y. Yu, Y. S. Wang, P. Huang and F. Weng, J. Phys. Chem. C, 2009, 113, 6406–6410.
10. F. Lahoz, Opt. Lett., 2008, 33, 2982–2984.
11. F. Liu, E. Ma, D. Q. Chen, Y. Yu and Y. S. Wang, J. Phys. Chem. B, 2006, 110, 20843–20846.
12. Y. Kawamoto, R. Kanno and J. Qiu, J. Mater. Sci., 1998, 33, 63–67.
13. A. De Pablos Martín, M. O. Ramírez, A. Durán, L. E. Bausá and M. J. Pascual, Opt. Mater., 2010, 33, 180–185.
14. X. M. Li, H. Guo, Y. Wei, Y. Guo, H. Lu, H. M. Noh and J. H. Jeong, J. Lumin., 2014, 152, 168–171.
15. S. Jiang, P. Zeng, L. Q. Liao, S. F. Tian, H. Guo, Y. H. Chen, C. K. Duan and M. Yin, J. Alloys Compd., 2014, 617, 538–541.
16. X. M. Li, J. K. Cao, Y. L. Wei, Z. R. Yang and H. Guo, J. Am. Ceram. Soc., 2015, 98, 3824–3830.
17. Y. L. Wei, X. Y. Liu, X. N. Chi, R. F. Wei and H. Guo, J. Alloys Compd., 2013, 578, 385–388.
18. S. Jiang, H. Guo, X. T. Wei, C. K. Duan and M. Yin, J. Lumin., 2014, 152, 195–198.
19. C. X. Li, S. Q. Xu, R. G. Ye, S. L. Zhao, D. G. Deng and S. L. Zhuang, Chin. Opt. Lett., 2010, 8, 66–69.
20. P. F. Li, M. Y. Peng, L. Wondraczek, Y. Q. Zhao and B. Viana, J. Mater. Chem. C, 2015, 3, 3406–3415.
21. M. Y. Peng, D. P. Chen, J. R. Qiu, X. W. Jiang and C. S. Zhu, Opt. Mater., 2007, 29, 556–561.
22. R. L. Leonard, A. R. Lubinsky and J. A. Johnson, J. Am. Ceram. Soc., 2017, 100, 1551–1560.
23. J. K. Cao, W. P. Chen, L. P. Chen, X. Y. Sun and H. Guo, Ceram. Int., 2016, 42, 17834–17838.
24. J. P. Zhang, D. C. Yu, F. F. Zhang, M. Y. Peng and Q. Y. Zhang, Opt. Mater. Express, 2014, 4, 111–120.
25. J. K. Cao, X. M. Li, Z. X. Wang, Y. L. Wei, L. P. Chen and H. Guo, Sens. Actuators, B, 2016, 224, 507–513.
26. G. Krieke, A. Sarakovskis and M. Springis, J. Alloys Compd., 2017, 694, 952–958.
27. Y. Liu, D. Tu, H. Zhu, R. Li, W. Luo and X. Chen, Adv. Mater., 2010, 22, 3266–3271.
28. C. Liu, H. Wang, X. Zhang and D. Q. Chen, J. Mater. Chem., 2009, 19, 489–496.
29. J. Xu, S. Gai, P. a. Ma, Y. Dai, G. Yang, F. He and P. Yang, J. Mater. Chem. B, 2014, 2, 1791–1801.
30. J. Yang, W. Huang, Y. Cheng, C. Wang, Y. Zhao, L. Zhu and X. Cao, CrystEngComm, 2012, 14, 899–907.
31. A. Herrmann, M. Tylkowski, C. Bocker and C. Rüssel, Chem. Mater., 2013, 25, 2878–2884.
32. D. Q. Chen, Z. Y. Wan, Y. Zhou, Y. Chen, H. Yu, H. W. Lu, Z. G. Ji and P. Huang, J. Alloys Compd., 2015, 625, 149–157.
33. J. B. Zhao, X. L. Zheng, E. P. Schartner, P. Ionescu, R. Zhang, T. L. Nguyen, D. Y. Jin and H. Ebendorff-Heidepriem, Adv. Opt. Mater., 2016, 4, 1507–1517.
34. V. K. Tikhomirov, L. F. Chibotaru, D. Saurel, P. Gredin, M. Mortier and V. V. Moshchalkov, Nano Lett., 2009, 9, 721–724.
35. H. Guo, N. Dong, M. Yin, W. Zhang, L. Lou and S. Xia, J. Phys. Chem. B, 2004, 108, 19205–19209.
36. B. R. Reddy, I. Kamma and P. Kommidi, Appl. Opt., 2013, 52, B33–B39.
37. K. Z. Zheng, Z. Y. Liu, C. J. Lv and W. P. Qin, J. Mater. Chem. C, 2013, 1, 5502–5507.
38. P. Du, L. H. Luo, W. P. Li, Q. Y. Yue and H. B. Chen, Appl. Phys. Lett., 2014, 104, 152902.
39. P. Du, L. H. Luo and J. S. Yu, RSC Adv., 2016, 6, 94539–94546.

---