Up-Conversion Photoluminescence in Thulia and Ytterbia Co-Doped Yttria-Stabilized Zirconia Single Crystals

Abstract

ZrO₂ is an attractive host matrix for luminescence material because of its excellent physical properties, such as low phonon energy and wide band gap. In this work, the highly transparent Tm₂O₃ and Yb₂O₃ co-doped yttria stabilized zirconia (YSZ) (abbreviated as Yb/Tm: YSZ) single crystals were grown by the optical floating zone method. The Yb/Tm: YSZ samples were stabilized in the cubic phase at room temperature when Yb³⁺ and Tm³⁺ replaced Y³⁺. The influence of Yb³⁺ co-doping on the up-conversion luminescence properties of the crystals was systematically studied. A total of 0.5 mol% Tm₂O₃ and 2.0 mol% Yb₂O₃ co-activated YSZ single crystal (abbreviated as 2.0Yb/Tm: YSZ) has the maximum luminous intensity. There were seven absorption peaks located at around 358, 460, 679, 783.3, 850–1000, 1200, and 1721.5 nm that were observed in the absorption spectrum of the 2.0Yb/Tm: YSZ single crystal. There were three up-conversion peaks at around 488, 658 and 800 nm that were observed when the Yb/Tm: YSZ samples were excitated at 980 nm. The fluorescence lifetime of Tm³⁺ for the ¹G₄→³H₆ transition of the 2.0Yb/Tm: YSZ sample is 7.716 ms as excited with a 980 nm laser. In addition, the oscillator strength parameters Ω_λ (λ = 2, 4 and 6) of this sample were derived by the Judd–Ofelt theory to evaluate the laser performance of the host materials. The ratio Ω₄/Ω₆ of this sample is 0.80, implying its excellent laser output. Therefore, the 2.0Yb/Tm: YSZ single crystal is a considerable potential material for laser and luminescence applications.

1. Introduction

Rare earth ions activated up-conversion luminescence material can convert near-infrared light into visible light by absorbing two or more low-energy photons and emitting one high-energy photon. This phenomenon violates the Stokes law, called the anti-Stokes luminescence or the up-conversion luminescence [1]. As they can emit short wavelength (high energy) light under the excitation of long wavelength (low energy) light source, the up-conversion luminescent materials have attracted great attention in the field of the infrared anti-counterfeiting, the anti-stokes cold light refrigeration, the up-conversion laser, the up-conversion 3D display, the sensing, and the medical applications [2,3,4,5,6,7].

As the emitted photon energy is greater than that of the excited ones, the energy levels between the activator and the sensitizer have to be very close to realize the continuous absorption of photons and energy transfer in the up-conversion luminescence process. Among the trivalent rare earth ions, Tm³⁺ is an important up-conversion luminescence activator. It has a ladder-like schema of energy level, and its ¹D₂→³F₄ and ¹G₄→³H₆ transition wavelengths are around 450 and 480 nm, respectively. The blue up-conversion luminescence process of Tm³⁺ can be excited under the pumping at 650, 800, and 980 nm. Tm³⁺ has been extensively studied due to its suitability for commercial InGaAs (940–990 nm) laser diode pumping. Yb³⁺ has the electronic configuration of 4f₁₃, and its energy state structure only contains one excited state(²F_5/2), which is slightly higher than the metastable excited state of Tm³⁺ (³H₅). Besides, Yb³⁺ has a strong absorption cross-section at 980 nm. The effective energy transfer can occur between Yb³⁺ and Tm³⁺, achieving efficient up-conversion luminescence. The co-doping of Yb³⁺ and Tm³⁺ can substantially improve the up-conversion luminescence efficiency [8,9,10,11,12], which was widely adopted in investigating and utilizing light radiation.

The efficiency of up-conversion luminescence material depends not only on the characteristics of rare earth ions, but also on the host material [13,14,15,16]. Seeking appropriate host materials to achieve high efficiency, high sensitivity, and a stable up-conversion luminescence laser remains a challenge. Among various inorganic host materials, fluorides were widely studied because of their lower phonon energy and higher up-conversion emission efficiency. However, their chemical and thermal instability limits the potential applications in different environments. Besides, the sensitivities of present optical thermal sensors decline rapidly with increasing temperature. Oxide materials have attracted extensive attention because of their stable physical and chemical properties, such as good thermal stability, oxidation resistance, high mechanical strength, and being green and pollution-free.

Among many oxides, zirconia (ZrO₂) is proven to be an excellent matrix material for the trivalent lanthanide ions doping due to its low phonon energy (470 cm⁻¹) [8], wide band gap (5.0 eV) [9], high density, low thermal expansion, and large chemical stability. However, the ZrO₂ has three phases, including monoclinic (lower than 1170 °C), tetragonal (1170 °C–2370 °C), and cubic (larger than 2370 °C) [9]. The phase transformation of ZrO₂ often comes with a volume change, leading to cracking. It is necessary to keep ZrO₂ stable in the cubic phase. One of the important ways is adding the stabilizers to ZrO₂ to form a stable solid solution. The more common way is adding 8 mol% Y₂O₃ to ZrO₂ to form the yttria zirconia solid (YSZ) solution, which has a cubic structure from room temperature to the melting point.

It is difficult to prepare high quality YSZ single crystal by conventional methods due to its high melting point (2700 °C). The optical floating zone technology can concentrate the light at the same point to increase the heating temperature to 3000 °C. Also, it can be used to grow metal oxide single crystals with a high melting point as no crucible is needed in the growth process. Our research group has successfully grown Y₃Al₅O₁₂ [17,18,19,20] and ruby [20,21] single crystals by this method. Therefore, in this work, the YSZ single crystals co-activated by various concentrations of Yb³⁺ and Tm³⁺ were grown by the optical floating zone technology. The structure and up-luminescence properties of the samples were characterized. Moreover, we utilized the Judd–Ofelt theory to analyze the spectral parameters of Tm³⁺ and Yb³⁺ in the samples.

2. Materials and Methods

2.1. Crystal Growth

A series of polycrystalline ceramic rods were prepared through a solid-state reaction process before the single crystal growth. The oxides ZrO₂ (99.99%), Y₂O₃ (99.99%), Tm₂O₃ (99.99%), and Yb₂O₃ (99.99%) powders (Aladdin, Shanghai, China) were employed as raw materials. The appropriate amounts of powders were weighed at specified mole ratios (as seen in

Table 1. Chemical compositions of Tm₂O₃ and Yb₂O₃ co-activated YSZ single crystals.

Samples	Composition (mol%)
	YSZ	Tm2O3	Yb2O3
1.0Yb/Tm: YSZ	98.5	0.5	1.0
2.0Yb/Tm: YSZ	97.5	0.5	2.0
3.0Yb/Tm: YSZ	96.5	0.5	3.0
4.0Yb/Tm: YSZ	95.5	0.5	4.0
5.0Yb/Tm: YSZ	94.5	0.5	5.0

) and mixed homogeneously by a magnetic stirrer for 24 h. Then, each mixture was oven-dried at 85 °C for 24 h and grounded in an agate mortar for 0.5 h. The prepared powder was packed in a long rubber balloon, vacuumed, sealed, and isostatically pressed under a pressure of 68 MPa to obtain the feed and seed rods. The obtained rods were sintered at 1550 °C for 10 h in the air to be compact and uniform polycrystalline ceramic rods.

Single crystals of Tm₂O₃ and Yb₂O₃ co-doped YSZ were grown using an optical floating zone furnace (FZ-T-12000-X-VII-VPO-GU-PC, Crystal Systems Co., Yamanashi, Japan). During growth, a counter-rotated seed of 10 rpm of the seed and feed rods, a flow rate of 4 L/min of air was maintained, and the crystal growth rate was about 5 mm/h.

The as-grown Yb/Tm: YSZ single crystals were colorless and about 5 × 80 mm in size, as shown in Figure 1. The crystals had a smooth surface and were free from cracks. The chips of about 1.0 mm in thickness were cut from the Yb/Tm: YSZ crystal rods and then double-sided polished.

2.2. Characterization

The X-ray diffraction (XRD) patterns were collected by a diffractometer (DX-2700, Haoyuan, Dangdong, China) using Cu-Kα (λ = 1.5406 Å) radiation in the 2θ range of 20°–80° with a step of 0.02°. The existing phases of as-prepared crystal chips were investigated also by Raman scattering. The Raman spectra in the wavenumber range of 150–950 cm⁻¹ were obtained by a confocal Raman spectrometer (inVia Reflex, Renishaw, London, UK) with the laser excitation at 532 nm.

The absorption spectra of crystal chips were collected on a UV-Visible-NIR spectro photometer (UV-3600, Shimadzu, Kyoto, Japan) with a resolution of 1 nm. The steady emission spectra were measured by a photoluminescence (PL) spectrometer (ZLF325, Zolix Instruments Co., Ltd, Beijing, China) with a 980 nm laser diode as an excitation source. Fluorescence decay curves were recorded with a fluorescence spectrometer (FLS920, Edinburgh Instruments, Edinburgh, UK) excited with a 980 nm laser as the excitation light.

3. Results and Discussion

3.1. Structure Analysis

The as-grown Yb/Tm: YSZ single crystals were crushed and ground into a powder to obtain detailed structure information. The XRD patterns of Yb/Tm: YSZ powders are shown in Figure 2a. For all the samples, six peaks were detected. It is impossible to identify the cubic and the tetragonal ZrO₂ as they share very similar X-ray reflection [12,22]. Raman spectroscopy is widely used to distinguish the three possible structures of ZrO₂ [23,24,25]. The Raman spectra of Yb/Tm: YSZ crystal chips are shown in Figure 2b. The Raman spectra of all samples are characterized by one strong peak in the wave-number range of 150–950 cm⁻¹. The Raman peak centered at around 625 cm⁻¹, corresponding to the F_2g mode of the cubic phase [26], is a little bigger than that of 0.5 Tm₂O₃:YSZ single crystal [12], and it may attributed to the presence of co-doped Yb₂O₃. The diffraction patterns of all the samples match well with the standard card of cubic ZrO₂ (JCPDS No.97-008-9429), belonging to the Fm-3m (225) space group. No extra peaks of a secondary phase were observed, indicating that Tm³⁺ and Yb³⁺ successfully entered the YSZ lattice. In addition, the cell parameters decrease slightly with the increase of Yb₂O₃ content, as seen in

Table 2. Cell parameters of x Yb/Tm: YSZ single crystals.

Samples	Cell Parameters (Å) a = b = c
1.0Yb/Tm: YSZ	5.1590
2.0Yb/Tm: YSZ	5.1581
3.0Yb/Tm: YSZ	5.1580
4.0Yb/Tm: YSZ	5.1555

. This is due to the Yb³⁺ preferentially substituting for Y³⁺, and the radius of Yb³⁺ (0.985 Å) is smaller than Y³⁺ (1.019 Å).

3.2. Absorption Spectra and Judd-Ofelt Analysis

The optical absorption spectra of x Yb/Tm: YSZ single crystals between 300 and 2000 nm were tested, as shown in Figure 3. There is no characteristic absorption related to d-d transition in the range of 300–2000 nm because the d orbit of Zr⁴⁺ is empty. Therefore, the absorption peaks of all the samples are from the transitions between Tm³⁺ and Yb³⁺. There are four visible and three IR absorption peaks, centering at 358, 460, 679, 783.5, 1200, and 1721.5 nm which correspond to the transitions of Tm³⁺: ³H₆→¹D₂, ³H₆→¹G₄, ³H₆→³F_2,3, ³H₆→³H₄, ³H₆→³H₅, and ³H₆→³F₄, respectively. The strongest absorption peak is at around 850–1000 nm, corresponding to the transition of Yb³⁺: ²F_7/2→²F_5/2, which indicated that Yb³⁺ has a large absorption of the 980 nm laser excitation and is suitable as a sensitizer in the up-conversion luminescence of Tm³⁺.

Optically, the Judd–Ofelt (J-O) theory, through an analysis of the oscillator strengths, the emission branching radio and the radiative lifetime [27,28,29] is used for analyzing the possible transition mechanisms of the rare earth ions that are affected by the host materials.

Based on the absorption spectrum of 2.0Yb/Tm: YSZ single crystal chip, the oscillator strengths for an induced electric dipole transition from the initial state to the final state are calculated according to the Equation (1) [28,29].

$$f=\frac{8{\pi }^{2}mc\sigma }{3h\left(2J+1\right)}\frac{{{\displaystyle \left({{\displaystyle n}}^2+2\right)}}^2}{9n}{\displaystyle {\sum }_{\lambda =2,4,6}{\mathsf{\Omega }}_{\lambda }}{{\displaystyle 〈4f^N\mathsf{\Psi }J//U^{\lambda }//4f^N{\mathsf{\Psi }}^{\prime }{J}^{\prime }〉}}^2$$

(1)

where m, c, σ, h, J, and n are the electron mass, velocity of light, transition energy, Planck constant, total angular momentum, and index of refraction, respectively. Ω_λ (λ = 2, 4 and 6) is oscillator intensity parameter and U_λ (λ = 2, 4 and 6) is the reduced matrix element. The values of average wavelength and the measured and calculated oscillated strengths for the 2.0Yb/Tm: YSZ single crystal are collected in

Table 3. Values of average wavelength and the experimental and calculated oscillated strengths for the 2.0Yb/Tm: YSZ single crystal.

3H6	$\overline{\mathsf{\lambda }}$ (nm)	Doubly Reduced Matrix Elements			Oscillator Strengths (10−20 cm2)
		‖U2‖2	‖U2‖4	‖U2‖6	Fexp	Fcal
1D2	359.62	0	0.3156	0.0928	0.0556	0.0533
1G4	482.73	0.0483	0.0748	0.0125	0.0914	0.0309
3F2 + 3F3	679.96	0	0.3164	1.0922	0.2338	0.2093
3H4	781.75	0.2373	0.1090	0.5947	0.1957	0.2028
3H5	1193.25	0.1074	0.2314	0.6383	0.2802	0.1714
3F4	1688.91	0.5375	0.7261	0.2382	0.3893	0.3465
RMS ΔS (10−20 cm2)	0.077

.

According to the J-O theory, the absorption oscillator strengths of Tm³⁺ ion are calculated, as shown in Table 3. The Judd–Ofelt parameters of Tm³⁺ in different host materials are listed in

Table 4. Judd–Ofelt parameters for Tm³⁺ doped in different hosts.

Crystal	Ω2 (10−20 cm2)	Ω4 (10−20 cm2)	Ω6 (10−20 cm2)	Ω4/Ω6	Ref
LLF	0.93	1.19	1.75	0.68	[31]
YVO4	9	1.05	2.7	0.389	[32]
SrWO4	7.41	0.25	0.98	0.15	[33]
YAG	0.7	1.2	0.5	2.4	[34]
YAP	0.67	2.30	0.74	3.108	[34]
YLF	2.42	1.28	0.90	1.422	[35]
SrGdGa3O7	1.29	1.08	0.47	2.3	[36]
YSZ	0.41	0.12	0.15	0.80	This work

. The root mean square (RMS) is 0.077 × 10⁻²⁰ cm², indicating the high reliability of the obtained oscillator strengths. The intensity parameters Ω_λ (λ = 2, 4, 6) of Tm³⁺ are 0.41 × 10⁻²⁰ cm², 0.12 × 10⁻²⁰ cm², and 0.15 × 10⁻²⁰ cm². It is well known that the Ω₂ parameter is sensitive to crystal structure and related to the covalency of the RE³⁺ sites. While Ω₄ and Ω₆ depend on the viscosity and rigidity of the host material. The value of Ω₂ is larger than that of Ω₄, revealing the relatively low symmetry of the Tm³⁺ site and the covalence that exists between the Tm³⁺ ions and anions as well as the asymmetry around the metal ion site [30]. According to the previous analysis, oxygen vacancies can be formed when the Y³⁺, Tm³⁺ and Yb³⁺ ions occupy the sites of the small Zr⁴⁺. Based on the theoretical calculation, the oxygen vacancies are located at the nearest neighbor of Zr⁴⁺, leading to the seven-fold coordination and the symmetry reduction of the Zr⁴⁺ sites. It could be concluded that the Tm³⁺ occupy the Zr⁴⁺ site with seven coordination after entering YSZ lattice. Meanwhile, Ω₄/Ω₆ is one of the important parameters for the spectral characterization of the host materials. The value of Ω₄/Ω₆ for 2.0Yb/Tm: YSZ single crystal is 0.8, which is larger than 0.68, 0.389, and 0.15 in LLF [31], YVO₄ [32], and SrWO₄ [33], revealing that the sample is a potential material for laser output.

Using the Ω_λ values, the radiative properties such as spontaneous transition probability A_ed, branching ratios β, and radiative lifetime τ_rad for the transitions of Tm³⁺ in 2.0Yb/Tm: YSZ single crystal are calculated and listed in

Table 5. Calculated radiative transition rates, branching ratios, and radiative lifetimes for different transition levels of 2.0Yb/Tm: YSZ single crystal.

Initial Level	Final Level	$\overline{\mathsf{\lambda }}$ (nm)	Aed	β	τrad (ms)
1G4	3F2	1634	3.193	0.56	1.764
	3F3	1494	14.006	2.47
	3H4	1177	59.520	10.50
	3F4	763	210.582	37.15
	3H5	666	42.751	7.54
	3H6	488	236.809	41.78
3H4	3H5	2166	4.981	1.40	2.819
	3F4	1432	27.057	7.63
	3H6	801.8	322.641	90.97

.

3.3. Luminescence Properties

Figure 4 shows the emission spectra of the YSZ: 0.5 mol%Tm₂O₃, x mol%Yb₂O₃ (x = 1.0, 2.0, 3.0, 4.0, 5.0) in the wavelength range of 400–900 nm under the 980 nm laser excitation at room temperature. Clearly, under the excitation of 980 nm, the bright blue luminescence was observed by the naked eye, as shown in Figure 5. The emission spectra consisted of three main emission bands at around 488 (blue), 658 (red), and 799 nm (NIR), corresponding to the of ¹G₄→³H₆, ¹G₄→³F₄, and ³H₄→³H₆ transitions of Tm³⁺, respectively. Among these, the NIR emission corresponding to ³H₄→³H₆ at 799 nm is the dominant one, and is consistent with the literature. As the Yb₂O₃ concentration increases, the position and shape of the emission peaks is similar, indicating the same surrounding of Tm³⁺. In other words, the structure of YSZ crystals does not change significantly after Yb₂O₃ co-doping. The structure of the crystal affected the luminescence properties directly. The blue up-conversion luminescence of Tm³⁺ in cubic ZrO₂ is composed of three peaks, while it is a single peak in monoclinic ZrO₂. The blue emission band in Figure 4 is mainly composed of three peaks, indicating that the crystal structure is cubic, and it is consistent with the XRD and Raman spectroscopy results.

Figure 6 shows the intensities of the emission peaks of the samples. The luminescence intensity increases with an increase in Yb₂O₃ concentration up to 2.0 mol%, and then decreases dramatically, due to the concentration quenching effect. According to the Blasse theory, the process of non-radiative energy transfer depends on the radiative reabsorption or the electric multipolar interaction of the activated ion. However, the radiation reabsorption process only occurs in the case of a large overlap between the excitation spectrum of the sensitizer and the emission spectrum of the activator. As can be seen in Figure 4, the emission peaks of Tm³⁺ are located at 488, 658, and 799 nm, while the absorption peak of the Yb³⁺ is located at around 980 nm, and the peak overlap is very small. Therefore, the concentration quenching in Yb/Tm: YSZ single crystals is induced by the electric multipolar interaction.

To examine the up-conversion mechanisms, the emission spectra of (0.5 mol%Tm₂O₃, 2.0 mol%Yb₂O₃): YSZ single crystal under 980 nm excitation at different power are plotted in Figure 7. The luminescence intensity (I) increases with the power (P), following the multiphoton formula, I ∝ Pⁿ. Here, n is the number of photons to populate the emitting states and is obtained by the slope of the linear fits of lg I-lg P curves. The slopes are 3.13, 3.30, and 2.32 for the blue emission centered at 488 nm, red emission centered at 658 nm, and NIR emission centered at 799 nm, respectively. This indicates that the red and blue emissions involve three photons while the NIR emission needs two photos. Although a photo avalanche (PA) is a possible mechanism for up-conversion luminescence, it is excluded because no inflection was observed in the power curve. Therefore, the up-conversion luminescence of Yb/Tm: YSZ single crystal is realized by the energy transfer (ET), the ground state absorption (GSA), and the excited state absorption (ESA) processes.

Figure 8 is the energy level diagram of the Yb³⁺-Tm³⁺ co-doped system. The transition processes in this up-conversion luminescence can be explained by the following. First, the electron of Yb³⁺ is excited from the ground state ²F_7/2 to the ²F_5/2 energy level by absorbing one 980 nm photon. Then, the Yb³⁺ transfers the energy to the ground state of neighboring Tm³⁺ according to the ET process. After that, the electron in the ³H₆ state of Tm³⁺ is excited to the ³H₅ state via the GSA process. The ³F₄ state is populated by the fast non-radiative relaxation. Following this, the Yb³⁺ absorbs the second photon and transfers the energy to the Tm³⁺, leading to the excitation of the electron from the ³F₄ state of Tm³⁺ to the ³F_2,3 state via the ESA channel. Subsequently, the non-radiative transition between two activator ions will populate the ³H₄ state. Finally, the Yb³⁺ absorbs the third photon and transfers the energy to the ³H₄ state, leading to the electron excitation to the ¹G₄ state. Then, the up-conversion luminescence is generated by the transitions from the high to the low-energy states. Therefore, the emission of blue (488 nm), red (658 nm), and NIR (799 nm) are observed according to the ¹G₄→³H₆, ¹G₄→³F₄, and ³H₄→³H₆ transitions, respectively.

3.4. Luminescence Decay Kinetics

Figure 9a presents the fluorescence decay curve of the emission line ¹G₄→³F₄ (658 nm) of Tm³⁺ in x Yb/Tm: YSZ crystal (x = 1.0, 2.0, 3.0 mol%), excited by 980 nm at room temperature. From a single exponential function fitting, the average fluorescence lifetime τ of the ¹G₄→³F₄ transition of the x Yb/Tm: YSZ crystal (x = 1.0, 2.0, 3.0 mol%) is equal to 0.949 ms, 0.972 ms, and 0.953 ms, respectively. The 2.0Yb/Tm: YSZ single crystal has the longest lifetime. As the concentration of Tm³⁺ remains the same, the excited state lifetime of Tm³⁺ is affected by the energy transfer process between Yb³⁺ and Tm³⁺. In addition, the fluorescence decay curve for Tm³⁺ for the ¹G₄→³H₆ (488 nm) transition in 2.0Yb/Tm: YSZ single crystal is shown in Figure 9b. The average fluorescence lifetime is 7.716 ms. The decay time is larger than that of Tm: LGYSO [37] (1.669 ms), Tm: YVO₄ [38] (1.9 ms), Tm: SSO [39] (1.14 ms), revealing the suitability of the YSZ crystal as a host material for luminescence of Tm³⁺.

3.5. Color Chromaticity Coordinates

The color chromaticity characteristics of the Yb/Tm: YSZ single crystals were analyzed by the Commission International de l’éclairage (CIE) chromaticity coordinates diagram as shown in Figure 10 and

Table 6. Values of CIE coordinates calculated for the various samples.

Sample	CIE x	CIE y
1.0Yb/Tm: YSZ	0.1080	0.2661
2.0Yb/Tm: YSZ	0.1164	0.2521
3.0Yb/Tm: YSZ	0.1116	0.2528
4.0Yb/Tm: YSZ	0.1057	0.2576

. The color points of all the samples are in cyan region, and the purity of the blue emission is 79.5% for the 2.0Yb/Tm: YSZ single crystal.

4. Conclusions

High-quality Yb³⁺/Tm³⁺ co-doped YSZ cubic single crystals were grown by the optical floating zone method. With an increasing concentration of Yb₂O₃, the cell parameter decreases due to the Yb³⁺ preferential substitution for Y³⁺. There were seven absorption peaks that were observed in the range of 300–2000 nm, corresponding to the transitions from Tm^{3+ 3}H₆ ground state to the ¹D₂ (358 nm), ¹G₄ (460 nm), ³F_2,3 (679 nm), ³H₄ (783.3 nm), ³H₅ (1200 nm), and ³F₄ (1721.5 nm) excited states and that in the 850–1000 nm region of Yb³⁺ (²F_7/2→²F_5/2) transition. Based on the absorption spectrum, the oscillator strength parameters Ω_λ (λ = 2, 4, 6) were calculated by the Judd–Ofelt theory. The Ω₂ value is larger than Ω₄, and it indicates that Tm³⁺ preferentially substituted for Zr⁴⁺ with seven coordination. There were three up-conversion emission peaks around 488, 658, and 799 nm that were observed when Yb/Tm: YSZ single crystal was excited with a 980 nm laser. The intensities of the emission peaks reached maximum when the concentration of Yb₂O₃ was 2.0 mol%, and then the concentration quenching effect appeared due to the electric multipole interaction. In addition, a two-photon process is involved for populating the ³H₄ level of Tm³⁺, whereas a three-photon process is involved in the generation of blue emission. The energy transfer between Yb³⁺ and Tm³⁺ was confirmed by the changes of the lifetime of Tm³⁺ from the fluorescence decay, which also reached a maximum with 2.0 mol% Yb₂O₃. The fluorescence lifetime of Tm³⁺ for ¹G₄→³H₆ transition was measured to be 7.716 ms for 2.0Yb/Tm: YSZ single crystal excited with a 980 nm laser. Furthermore, the color coordinates of the 2.0Yb/Tm: YSZ sample had a blue luminescence purity of 79.5%. Therefore, the 2.0Yb/Tm: YSZ single crystal is a potential candidate for the laser and luminescence applications.