Up-conversion luminescence, temperature sensing properties and laser-induced heating effect of Er³⁺/Yb³⁺ co-doped YNbO₄ phosphors under 1550 nm excitation

Abstract

YNbO₄ phosphors with various Er³⁺ and Yb³⁺ concentrations were synthesized via a traditional high-temperature solid-state reaction method. Their crystal structure was investigated by means of X-ray diffraction (XRD) and Rietveld refinements, and it was confirmed that the obtained samples exist in monoclinic phase. The Er³⁺ and Yb³⁺ concentration-dependent up-conversion (UC) luminescence was studied under 1550 nm excitation. By inspecting the dependence of UC intensity on the laser working current, it was found that four-photon and three-photon population processes were co-existent for generating the green UC emissions in the samples with higher Yb³⁺ concentrations. In addition, it was observed that the temperature sensing properties of YNbO₄: Er³⁺/Yb³⁺ phosphors were sensitive to both Er³⁺ and Yb³⁺ doping concentrations. Furthermore, based on the obtained temperature response of the UC luminescence phosphors, 1550 nm laser-irradiation-induced thermal effect was studied, and it was discovered that the sample temperature was very sensitive to the doping concentrations of Er³⁺ and Yb³⁺ and the excitation power.

Introduction

Rare-earth (RE) ion-doped up-conversion (UC) luminescence materials have received considerable attention due to their widespread applications in many fields, such as UC lasers, sensors, solar cells, three dimension display and so on^1,2. In practical applications, the UC phosphors with high luminescence efficiency and reliable temperature sensing ability may be preferable. Nevertheless, the UC materials possessing both highly efficient UC efficiency and high temperature sensitivity are very fewer³. Therefore, it is necessary to explore a new type material to meet this requirement.

On the one hand, choosing proper trivalent RE ions used as absorption and emission centers in the design of UC luminescence materials is very important in order to obtain highly efficient UC emissions^4,5,6. Among trivalent RE ions, Er³⁺ is one of the most preferable RE ions for UC luminescence due to its uniform energy level distribution and proper energy distance between some metastable energy levels, which allows Er³⁺ to be effectively excited by different laser wavelengths, such as 800, 980 and 1550 nm⁷. Moreover, Er³⁺ is expected as an effective temperature sensing unit with high temperature sensing sensitivity based on its two thermally coupled emitting energy levels ²H_11/2 and ⁴S_3/2, whose energy distance is about 770 cm^−1 8,9. In order to further enhance the UC luminescence efficiency of Er³⁺, Yb³⁺ is often employed as a sensitizer owing to its large absorption cross-section at around 980 nm and efficient energy transfer to Er^3+ 10. Thus, Er³⁺/Yb³⁺ co-doped UC luminescence phosphors have been extensively studied^4,5,6,8,9. On the other hand, the characteristics of the host materials, such as their phonon energy, chemical and thermal stabilities, have significant effects on the UC luminescence properties of the UC phosphors^11,12,13,14. In general, oxide compounds have good chemical and physical stabilities, which are beneficial to the applications in high temperature environments, have received increasing attention in recent years^{15,16,17,18,19,20,21,22}. Among various oxide matrixes, ANbO₄ (A = Gd, Y, La) compounds, with a fergusonite structure and low phonon energy, are good host candidates for UC luminescence phosphors^{17,18,19,20,21,22}. Recently, the UC luminescence properties of Er³⁺/Yb³⁺ co-doped ANbO₄ materials under 980 nm excitation have been reported^17,18. It was found that the ANbO₄ host presented high chemical and physical stabilities, low lattice vibration energy and high accommodating concentration for RE luminescent centers; additionally they can be easily prepared via the traditional solid-state reaction method. However, the UC luminescence properties reported in these niobate systems are mainly focused on the 980 nm excitation. In addition to 980 nm excitation, the Er³⁺ can also be excited by 1550 nm. 1550 nm laser has obtained considerable attention due to its eye-safe nature and wide applications in optical communication and solar cells; meanwhile, it is also an effective excitation source to investigate the UC luminescence properties of RE³⁺-doped materials^23,24,25. It is reported that efficient UC emission from Er³⁺ was obtained under monochromatic excitation around 1500 nm²⁴. Previously, we have studied the UC luminescence properties of Er³⁺ in Er³⁺ mono-doped YNbO₄ phosphors under 1550 nm excitation²². The results showed that the relative intensity of the green and the red UC emissions were significantly different in down-conversion from that in UC luminescence processes because of the different population routes and pumping conditions. Furthermore, it was confirmed that energy transfer process was the main mechanism for UC emissions of Er³⁺ in Er³⁺ mono-doped YNbO₄ phosphors excited at 1550 nm. In addition, we have also studied the effect of Mo⁶⁺ on the UC luminescence and temperature sensing sensitivity of Er³⁺ in ANbO₄ phosphors excited at 1550 nm²⁰. However, there is little work on the study of UC luminescence, temperature sensing property and laser-induced thermal effect of Er³⁺/Yb³⁺ co-doped ANbO₄ phosphors under 1550 nm excitation.

Hereby, in this study, Er³⁺/Yb³⁺ co-doped YNbO₄ phosphors were prepared by a traditional high-temperature solid-state reaction method. Er³⁺ and Yb³⁺ concentration-dependent UC luminescence processes under 1550 nm excitation were discussed. Different concentration dependent UC luminescence processes for the green and red UC emissions were discovered. In addition, the effects of Er³⁺ and Yb³⁺ concentrations on the optical temperature sensing properties were also investigated according to fluorescence intensity ratio (FIR) technique. Furthermore, the laser-irradiation-induced thermal effect was evaluated by persistent irradiation of the 1550 nm laser. It was observed that both the temperature sensing property and laser-induced thermal effect were dependent on the Er³⁺ and Yb³⁺ concentrations.

Experimental Details

Sample preparation

Usually, doping concentration has significant effect on the luminescent performance of the UC luminescence materials. Previously, we have investigated the spectroscopic properties of Er³⁺ mono-doped YNbO₄ phosphor and obtained the optimum Er³⁺ concentration (5 mol%)²². According to our previous results^16,25, two sets of YNbO₄: Er³⁺/Yb³⁺ phosphors were prepared by a traditional high-temperature solid-state reaction method. In one set, the concentration of Er³⁺ was fixed to be 5 mol%, and the doping concentrations of Yb³⁺ were designed to be 0, 2, 6, 10, 14, 16 and 20 mol%. In the other set, the concentrations of Er³⁺ were respectively set as 0.5, 1, 3, 5, 7, 9, 12 and 15 mol%, and the concentration of Yb³⁺ was fixed to be 6 mol%, which was the optimum doping concentration derived from the experimental results of the first set of the samples.

The starting materials Y₂O₃ (99.99%), Yb₂O₃ (99.99%) and Er₂O₃ (99.99%) used in this work were purchased from Shanghai Second Chemical Reagent Factory (China), Nb₂O₅ (99.99%) were purchased from Tianjin Reagent Chemicals Co., Ltd (China). They were first weighed according to the designed stoichiometric ratio and fully ground in an agate mortar for 30 min until the mixtures were uniformly mixed. Furthermore, the mixture was transferred into a crucible, and then it was put into a muffle furnace (SX-GO 3163, Tianjin Zhonghuan Experimental Electric Furnace Co., Ltd., China) and calcined at 1300 °C for 4 h. Last, the final products were obtained after the furnace was naturally cooled to room temperature.

Sample characterization

A SHIMADZU X-ray diffractometer (XRD-6000) with CuKα₁ radiation (λ = 0.15406 nm) was used to analyze the crystal structure of those as-prepared samples. Temperature dependent UC luminescence spectra in the visible light region were measured by a Hitachi F-4600 fluorescent spectrometer equipped with a power-controllable 1550 nm fiber laser as an excitation source. The output power of the 1550 nm fiber laser shows a linear dependence on its working current. The fluorescence spectra in the wavelength range of 800–1100 nm were obtained by a near-infrared spectrometer (NIR QUEST-256) with an external 1550 nm fiber laser. The temperatures of the samples were controlled by a self-made sample temperature controlling system, DMU-TC 450, with a controlling accuracy of about ±0.5 °C.

Results and Discussion

Identification of the crystal structure

In order to identify the crystal structure of the as-prepared phosphors, XRD measurements were performed on all the samples. The results showed that all the samples almost had the same diffraction patterns. To better study the effect of Er³⁺, Yb³⁺ doping concentrations on the crystal structure of YNbO₄, a Rietveld refinement procedure was carried out with non-commercial software, named General Structure Analysis System (GSAS)^26,27. Figure 1 shows the XRD Rietveld refinement results of YNbO₄: x mol% Er³⁺/y mol% Yb³⁺ (x = 5, y = 0 and 20; y = 6, x = 9 and 15) phosphors as representatives. In the refinement processes, the crystallographic data of monoclinic YNbO₄ (space group of C2/c (15), see JPCDS card No. 83–1319) were used as the initial crystal structure model. The measured, calculated results, Bragg position, and the difference between experimental and calculated diffraction patterns are also shown in Fig. 1. It can be seen that all the diffraction peaks in the experimental data can be well fitted by the Rietveld theoretical model, and the deviation between the experimental and the calculated data are acceptable. The cell parameters and the Rietveld refinement results are shown in Table 1. It can be found that the lattice constants decrease with the increase of both Er³⁺ and Yb³⁺ concentrations, which confirms the disordered lattice induced by the substitution of larger Y³⁺ (0.90 Å) by smaller Er³⁺ (0.88 Å) and Yb³⁺ (0.86 Å). However, all the observed diffraction peaks for those samples satisfy the reflection conditions, and our as-prepared samples are of monoclinic phase.

Figure 1

XRD patterns of YNbO₄: x mol% Er³⁺/y mol% Yb³⁺ (x = 5, y = 0 and 20; y = 6, x = 9 and 15) phosphors. Asterisks represent the experimental data, the red curves accompanying the asterisks show the Rietveld refinement results, the green sticks mark the Bragg reflection positions, and the blue solid lines represent the differences between experimental and calculated data.

Table 1 Rietveld refinement results and crystal data for the YNbO₄: x mol% Er³⁺/y mol% Yb³⁺ (x = 5, y = 0 and 20; y = 6, x = 9 and 15).

Up-conversion luminescence properties

It is well known that the UC luminescence intensity greatly depends on the doping concentrations of RE ions, thus we have studied the effect of Er³⁺ and Yb³⁺ concentrations on the UC luminescence properties excited at 1550 nm wavelength. Er³⁺ and Yb³⁺ concentrations dependent UC luminescence spectra measured at room temperature are displayed in Fig. 2. Figure 2(a) shows the UC luminescence spectra of the as-prepared YNbO₄ phosphors with fixed Er³⁺ concentration of 5 mol% and various Yb³⁺ concentrations of 0, 2, 6, 10, 14, 16 and 20 mol%. Several characteristic emissions of Er³⁺ can be found in the UC luminescence spectra. Among them, two strong green emissions centered at around 545 and 565 nm can be assigned to the ²H_11/2, ⁴S_3/2 → ⁴I_15/2 transitions, and the red emission ranging from 650 to 740 nm originates from ⁴F_9/2 → ⁴I_15/2 transition^28,29. In addition, an extremely weak near infrared emission located at 820 nm corresponding to ⁴I_9/2 → ⁴I_15/2 transition can also be observed³⁰. The inset of Fig. 2(a) shows the dependences of the green and the red UC emission intensities on Yb³⁺ concentration. It can be seen that the concentration of Yb³⁺ has significant influence on both green and red UC emission intensities. With an increase in Yb³⁺ concentration, both green and red UC emission intensities gradually increase and reach their own maximum values when the concentration of Yb³⁺ is 6 mol%, and then the intensities decline as Yb³⁺ content further increases. Moreover, with increasing Yb³⁺ concentration the integrated intensity ratio (I _R /I _G ) of the red to the green UC emissions exhibits an interesting variation. As can be seen from the inset of Fig. 2(a), the intensity ratio I _R /I _G first decreases with an increase in Yb³⁺ concentration, and then significantly increases when the concentration of Yb³⁺ is higher than 6 mol%. The effect of Yb³⁺ concentration on the intensity ratio I _R /I _G will be discussed later.

Figure 2

UC luminescence spectra for YNbO₄ samples doped with various Yb³⁺ concentrations (a) and Er³⁺ concentrations (b) excited at 1550 nm. The insets show the dependences of the green and the red UC luminescence intensities and the intensity ratio (I _R /I _G ) of red to green UC emissions on Yb³⁺ and Er³⁺ concentrations.

To examine the effect of the concentration of Er³⁺ on the UC luminescence of YNbO₄: Er³⁺/Yb³⁺ under 1550 nm excitation, the UC luminescence spectra of those samples doped with 6 mol% Yb³⁺ and x mol% Er³⁺ (x = 0.5, 1, 3, 5, 7, 9, 12 and 15) were measured under an identical experimental conditions and are shown in Fig. 2(b). The intrinsic emissions of Er³⁺ as appearing in Fig. 2(a) are also observed. And when the concentration of Er³⁺ reaches 9 mol%, both green and red UC emission intensities are the strongest, and then decrease with further increase in Er³⁺ concentration, which can be seen in the inset of Fig. 2(b). In addition, from the inset of Fig. 2(b), it can also be found that the intensity ratio I _R /I _G decreases with an increase in Er³⁺ concentration in the whole of our studied concentration range, which will be discussed later, too.

In order to further analyze the UC luminescence processes of Er³⁺ in YNbO₄: Er³⁺/Yb³⁺ phosphors under 1550 nm excitation, the laser working current dependence of UC luminescence of YNbO₄: 5 mol% Er³⁺/y mol% Yb³⁺ (y = 0, 6, 14 and 20) and YNbO₄: x mol% Er³⁺/6 mol% Yb³⁺ (x = 0.5, 3, 9 and 15) phosphors were studied. As examples, the UC luminescence spectra for both YNbO₄: 5 mol% Er³⁺/6 mol% Yb³⁺ and YNbO₄: 6 mol% Yb³⁺/9 mol% Er³⁺ samples under 1550 nm excitation at different laser working current are shown in Fig. 3(a,b). It should be pointed out that these two samples display the strongest UC luminescence in each set of phosphors in this study. From Fig. 3, it can be seen that the UC luminescence intensities of both samples monotonically increase with an increase of the laser working current. Strong green and red UC emissions were both observed. To further clarify the UC luminescence mechanism of Er³⁺, the integrated intensities of the green and the red UC emissions are calculated separately, and the results for various concentrations of Yb³⁺ and Er³⁺ doped YNbO₄ phosphors are shown in Fig. 4(a,b). For any unsaturated UC luminescence process, the UC emission intensity (I _up ) is depended on the laser working current (i_LD), and fulfills the following relation³¹:

$${I}_{up}\propto {(a{i}_{{\rm{LD}}}-b)}^{n}$$

(1)

where n is the number of required photon, a and b are constants. Equation (1) is used to fit the experimental data depicted in Fig. 4(a,b), and the fitting results are also shown in Fig. 4(a,b). As can be seen, all these n values of the red UC emission (n_R) are around 3, indicating that the red UC emissions are dominated by three-photon process for different concentrations of Yb³⁺ and Er³⁺ doped samples. The little differences for those n_R values may be caused by the calculation error of the integrated intensity and the error in the fitting process. As can be seen in Fig. 4(a), for the green UC emissions the fitted n values (n_G) are also around 3. But the value of n_G continues to grow as Yb³⁺-concentration increases, and it is greater than 3 when Yb³⁺-concentration reaches 14 mol%, indicating much more 1550 nm photons are needed to achieve the green UC emissions in the case of high Yb³⁺-concentration.

Figure 3

UC luminescence spectra of YNbO₄: 5 mol% Er³⁺/6 mol% Yb³⁺ (a) and YNbO₄: 9 mol% Er³⁺/6 mol% Yb³⁺ (b) phosphors upon 1550 nm excitation at different laser working current.

Figure 4

Relationship between the integrated intensities of the green and the red UC emissions and laser working current for different samples, (a) 5 mol% Er³⁺/y mol% Yb³⁺ (y = 0, 6, 14 and 20); (b) x mol% Er³⁺/6 mol% Yb³⁺ (x = 0.5, 3, 9 and 15).

To better describe the concrete UC luminescence processes of Er³⁺ in YNbO₄: Er³⁺/Yb³⁺ phosphors excited at 1550 nm wavelength, Fig. 5 depicts the simplified energy level diagrams of Er³⁺, Yb³⁺ and the possible energy transfer processes. In principle, there are three possible basic population mechanisms involved in the UC luminescence processes, including energy transfer up-conversion (ETU), ground state absorption (GSA) and excited state absorption (ESA)¹⁰. As can be seen in Fig. 5, the UC luminescence of Er³⁺ under 1550 nm excitation can be obtained by both direct light absorption of Er³⁺ and ET from Yb³⁺. In this case, for Er³⁺ single-doped sample the population of ²H_11/2 and ⁴S_3/2 levels can be achieved by successive absorbing three 1550 nm photons via GSA and two ESA or ET (from Er³⁺ to Er³⁺) processes, and then yielding the green UC emissions^24,32. The possible UC luminescence process of the red UC emission is that Er³⁺ in ⁴I_9/2 level non-radiatively relaxes to ⁴I_11/2 level and absorbs one 1550 nm photon or receives an energy transferred from a neighboring Er³⁺, yielding the population of ⁴F_9/2 level, and then red UC emissions can be achieved.

Figure 5

UC luminescence mechanism models of Er³⁺ and possible ET and CR channels of YNbO₄: Er³⁺/Yb³⁺ phosphors under 1550 nm excitation.

Compared with Er³⁺ mono-doped YNbO₄ phosphors, introducing Yb³⁺ as a sensitizer can influence the UC luminescence process of Er³⁺ though Yb³⁺ cannot directly absorb the energy of 1550 nm photons. From the inset of Fig. 2(a), it can be found that Yb³⁺ concentration has significant effect on the UC luminescence intensity of Er³⁺. With an increase of Yb³⁺ concentration, the UC luminescence intensity of Er³⁺ first increases and then decreases, which may be caused by the back energy transfer (BET, ⁴I_11/2 (Er³⁺) + ²F_7/2 (Yb³⁺) → ⁴I_15/2 (Er³⁺) + ²F_5/2 (Yb³⁺)) process from Er³⁺ to Yb^3+ 15. Er³⁺ ions first absorb the energy of two 1550 photons to achieve the population of ⁴I_9/2 level, and then complete the population of ⁴I_11/2 level by non-radiative relaxation process. Because the energy distance between ²F_5/2 level and ²F_7/2 level of Yb³⁺ matches well with that of ⁴I_11/2 level and ⁴I_15/2 level of Er³⁺, energy transfer from Er³⁺ to Yb³⁺ can occur. The existence of the BET process can be confirmed by the UC luminescence spectra shown in Fig. 6(b), which were measured under 1550 nm excitation in the near infrared region. As can be seen, the UC emission intensity from Yb³⁺ increases first and then decreases with an increase in Yb³⁺ concentration. Considering the efficient ET process from Yb³⁺ to Er³⁺, Er³⁺ → Yb³⁺ → Er³⁺ ETU process should be an important mechanism for the UC luminescence of Er³⁺ when excited at 1550 nm^11,33,34. Therefore, the UC luminescence mechanism of Er³⁺ can be described by the following processes. For Er³⁺ and Yb³⁺ co-doped case, Er³⁺ first absorbs two 1550 nm photons populating the ⁴I_9/2 level and achieves the population of ⁴I_11/2 level via a following non-radiative relaxation process. Next, Er³⁺ transfers the energy to an adjacent Yb³⁺ via the BET process, and then Yb³⁺ is excited to its excited state ²F_5/2 level. Then, the excited Yb³⁺ transmits part energy to a neighboring Er³⁺, achieving the populations of ²H_11/2/⁴S_3/2 and ⁴F_9/2 levels of Er³⁺ via ET2 and ET1 processes, respectively. Finally, the green and the red UC emissions can be achieved. For the green UC emissions, the above-mentioned BET process needs to absorb two 1550 nm photons, and another two 1550 nm photons are needed to be absorbed to accomplish the population of ⁴I_11/2 level via ET2 process. Thus, four 1550 nm photons are needed for the green UC emission when Yb³⁺ acts as the sensitizer of Er³⁺. Because those values n_G derived from the fitting process are larger than 3 especially in the case of high concentration of Yb³⁺ doping, it can be concluded that three-photon and four-photon absorption processes co-existent in the green UC luminescence process. That is to say, ETU processes between Er³⁺ and Yb³⁺ and between Er³⁺ ions are co-existent in the green UC luminescence of Er³⁺ under 1550 nm excitation.

Figure 6

(a) Infrared UC luminescence spectra for YNbO₄: 6 mol% Yb³⁺/5 mol% Er³⁺, and the integrated infrared UC intensities of the samples doped with varied (b) Yb³⁺ and (c) Er³⁺ concentrations excited at 1550 nm wavelength.

Furthermore, the influences of Er³⁺ and Yb³⁺ concentrations on the UC luminescence intensity and the intensity ratio I _R /I _G can also be better explained by the energy level diagrams of Er³⁺ and Yb³⁺. For the case of varying Yb³⁺ concentration, when the concentration of Yb³⁺ is not higher than 6 mol%, much more energy can be transferred from Er³⁺ to Yb³⁺ with an increase of Yb³⁺ concentration, which means BET process occurs. Then, higher energy is re-transferred from Yb³⁺ to Er³⁺ via ET1 and ET2 processes, achieving the populations of the green and the red excited state levels ²H_11/2/⁴S_3/2 and ⁴F_9/2. However, the increase of the population of ⁴I_13/2 level will be suppressed compared with that of ⁴I_11/2 level due to the increased BET process. Finally, the UC luminescence intensity enhances, but the intensity ratio I _R /I _G of Er³⁺ decreases. In the case of higher concentration of Yb³⁺ (greater than 6 mol%) doping, the interionic distance between Er³⁺ and Yb³⁺ decreases and those aggregated Yb³⁺ may become quenching center, resulting in the quenching of UC luminescence of Er^3+ 35. In addition, at higher doping concentration of Yb³⁺, the population of ⁴I_11/2 level may be reduced via BET process much easier than that of ⁴I_13/2 (as can be seen in Fig. 6(b), the UC luminescence intensity of the emission from ⁴I_11/2 → ⁴I_15/2 transition decreases when the concentration of Yb³⁺ reaches 6 mol%), which further affects the populations of ²H_11/2/⁴S_3/2 and ⁴F_9/2 levels. Thus, the intensity of the green and the red UC emissions decrease and the ratio I _R /I _G increases with an increase of Yb³⁺ concentration.

For variable concentrations of Er³⁺-doped case, the UC luminescence intensity also increases first and then decreases with an increase in Er³⁺ concentration. At higher concentration of Er³⁺, the reduction of UC luminescence intensity of Er³⁺ can be ascribed to the concentration quenching, which may be caused by the non-radiative ET and the cross-relaxation (CR) between Er³⁺ due to the intensified interaction between Er³⁺ for the closer interionic distance. The possible CR processes are shown in Fig. 5²⁵. As can be seen, CR1-CR4 processes are responsible for the depopulation of the green emitting levels ²H_11/2 and ⁴S_3/2, but CR5 and CR6 processes take the responsibility for the depopulation of the red emission level ⁴F_9/2, which will cause the quenching of the corresponding green and red UC emissions, respectively. With an increase in Er³⁺ concentration, the BET process from Er³⁺ to Yb³⁺ can also be enhanced, and those aggregated Yb³⁺ may be one type of quenching center, which may further induce the quenching of the UC luminescence of Er^3+ 34. In addition, as mentioned above, the intensity ratio I _R /I _G decreases with an increase of Er³⁺ concentration in the whole of the studied concentration range, which indicates that the depopulation of the red emitting level ⁴F_9/2 is much faster than that of the green emitting levels ²H_11/2 and ⁴S_3/2. As can be seen from Fig. 5, CR1 process could increase the population of ⁴F_9/2 level, so this CR process can be ignored in the present case. The other CR processes may simultaneously influence the population of the green and the red emitting levels assisted by other ET, energy absorption or non-radiative transition processes.

Temperature sensing properties

Temperature is an important physical quantity in both science and industrial fields, and it has a great influence on the spectroscopic properties of the luminescent materials. However, it is difficult to monitor the actual temperature of the materials via traditional contact measurement due to the interference from the introduced thermometers. Thus, it is necessary to find a non-contact method to measure the internal temperature of one material.

The energy distance between the two levels ²H_11/2 and ⁴S_3/2 of Er³⁺ is small and these two levels exist in a state of thermal equilibrium, thus their populations fulfill the Boltzmann’s distribution³⁶. And then the FIR of these two green emissions is temperature dependent. Therefore, Er³⁺ can be used as temperature sensing unit to achieve temperature detection via probing the FIR of the two green emissions. According to the Boltzmann’s distribution theory, the dependence of the FIR value (R, I _H /I _S ) on temperature for Er³⁺-doped samples can be expressed as follows³⁷:

$$R(T)=\frac{{I}_{H}}{{I}_{S}}=A\,\exp (\frac{-{\rm{\Delta }}E}{kT})$$

(2)

where I _H and I _S are the integrated fluorescence intensities of the emissions from ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 transitions, respectively. ΔE is the energy distance between ²H_11/2 and ⁴S_3/2 levels, k is Boltzmann’s constant, and T is absolute temperature. To investigate the optical temperature sensing behavior of Er³⁺ in YNbO₄: Er³⁺/Yb³⁺ phosphors, the UC luminescence spectra of the samples doped with 5 mol% Er³⁺/y mol% Yb³⁺ (y = 0, 2, 6 and 20) and x mol% Er³⁺/6 mol% Yb³⁺ (x = 0.5, 1, 9 and 15) were measured at various temperatures ranging from 293 K to 723 K. Figure 7 shows the temperature dependent UC luminescence spectra for the samples with the maximum UC luminescence intensity of the two groups in our study. As can be seen, the two green emission bands originated from ²H_11/2, ⁴S_3/2 → ⁴I_15/2 transitions can be observed, and the overall luminescence intensity decreases with an increase of temperature, indicating temperature dependent luminescence quenching occurs. Generally, non-radiative relaxation, energy transfer and crossover process are the main three reasons which may cause the temperature quenching of luminescence^16,25. With an increase of sample temperature, the non-radiative relaxation rate of the emitting levels, the energy transfer probability originating from the emitting levels, and the crossover process between the emitting levels of the luminescent centers and the charge transfer band of them or the absorption band of the matrix in high energy region will be enhanced, which will result in the quenching of the luminescence intensity. Similar to our previous research results, the non-radiative relaxation process would be the main mechanism for the temperature dependent UC luminescence quenching of Er³⁺ in the present case^16,25. In addition, it can also be found that the peak positions of them almost unchanged but the relative intensity and the changing trend of each green UC emission are different with an increase of temperature. It can be seen that the intensity of the green UC emission originated from ²H_11/2 → ⁴I_15/2 transition first decreases slightly and then remains approximately unvaried. However, the intensity of the emission from ⁴S_3/2 → ⁴I_15/2 transition monotonically decreases with an increase of temperature. However, the intensity ratios R monotonically increases with increasing the sample temperature, which can be seen from Fig. 8(a,b). As can be seen in Fig. 8(a,b), all the experimental data can be fitted well by equation (2). The confirmed equations for different samples are also shown in Fig. 8(a,b). As can be seen, all of the values of ΔE/k are around 1000. The corresponding ΔE value was 694.5 cm⁻¹, which is in good agreement with the value (713.8 cm⁻¹) derived from the measured UC luminescence spectra in Fig. 2. This fact means that equation (2) can explain well the temperature dependence of the FIR of the two green UC emissions.

Figure 7

Temperature dependent UC luminescence spectra for (a) YNbO₄: 5 mol%Er³⁺/6 mol% Yb³⁺ and (b) 6 mol% Yb³⁺/9 mol% Er³⁺ under 1550 nm excitation.

Figure 8

The relationships between the intensity ratios R and the sample temperature for YNbO₄ phosphors doped with 5 mol% Er³⁺/y mol% Yb³⁺ (y = 0, 2, 6 and 20) (a) and x mol% Er³⁺/6 mol% Yb³⁺ (x = 0.5, 1, 9 and 15) (b).

With regard to the application in optical temperature sensing, it is extremely important to be aware of the temperature sensing sensitivity. Usually, the sensor sensitivity can be defined as the change rate of R per unit temperature and can be expressed as follows³⁸:

$$S(T)=\frac{dR(T)}{dT}=A\,\exp (\frac{-{\rm{\Delta }}E}{kT})(\frac{{\rm{\Delta }}E}{k{T}^{2}})$$

(3)

By taking the values of A and ΔE/k derived from the fitting processes in Fig. 8(a,b) into equation (3), the sensitivities of the samples were calculated and are shown in Fig. 9(a,b). It can be seen that the sensitivities of all the samples have the same changing trend, and all of them are dependent on the sample temperature. With an increase of sample temperature, the sensitivity first increases and reaches the maximum value at around 536 K, and then dramatically decreases. In addition, it can also be found that the concentrations of Er³⁺ and Yb³⁺ have a great effect on the temperature sensing sensitivity. The sensitivity decreases with an increase of the doping concentrations of Er³⁺ and Yb³⁺. Similar phenomenon were observed in YNbO₄: Er³⁺ and NaGdTiO₄: Er³⁺, Yb³⁺ phosphors in our previous studies, which was attributed to the different optical transition rate of Er³⁺ in different samples doped with various concentrations of Er^3+ 22,39. Therefore, if we want to obtain high temperature sensitivity, the doping concentrations of Er³⁺ and Yb³⁺ should be further reduced. However, when the concentration is too low, the luminescence of the sample is too weak, which is likely to cause large spectral measurement errors. Therefore, we still need to select the doping concentration of RE ions properly in practical applications. In our present study, the maximum sensitivity of Er³⁺ is 83.80 × 10⁻⁴ K⁻¹ for 0.5 mol% Er³⁺/6 mol% Yb³⁺ doped sample.

Figure 9

Sensitivity curves for YNbO₄: Er³⁺/Yb³⁺ phosphors doped with 5 mol% Er³⁺ and 0, 2, 6, 20 mol% Yb³⁺ (a); 6 mol% Yb³⁺ and 0.5, 5, 9, 15 mol% Er³⁺ (b).

Laser-induced thermal effect

The UC luminescence process is often accompanied by heat generation, especially when the luminescence materials are excited at long wavelength laser source with high power density. The laser irradiation can result in an elevation of the sample temperature, and further bring temperature dependent effect on the spectroscopic properties of the luminescence materials, thus it is necessary to study the laser-induced thermal effect of the UC luminescence materials.

To understand the thermal effect induced by 1550 nm laser irradiation on the samples, the UC emission spectra of YNbO₄: Er³⁺/Yb³⁺ phosphors with various doping concentrations of Er³⁺ and Yb³⁺ were measured when these samples were continuously irradiated by 1550 nm laser with working current of 2.00A for 10 minutes. The integrated intensity ratios of the two green emissions were calculated according to the UC luminescence spectra, and the sample temperatures were derived by taking the integrated intensity ratios into equation (1). Figure 10(a,c) show the dependence of the sample temperature on the irradiation time of 1550 nm laser for the Er³⁺ and Yb³⁺ concentration-varied samples, the working current of 1550 nm laser is fixed at 2.00 A. It can be seen that the temperatures of all the samples change very slightly with irradiation time, indicating that all the samples can quickly reach thermal equilibrium. In addition, it can also be found that the sample temperatures are higher than the room temperature, suggesting the existence of laser-induced thermal effect. Figure 10(b,d) show the dependence of the sample temperatures on the working current of 1550 nm laser for the Er³⁺ and Yb³⁺ concentration-varied samples. It can be seen that the sample temperature increases with an increase of the laser working current, indicating significant laser-induced thermal effect in high power density (corresponding to large working current of 1550 nm laser) pumped case. Moreover, it is also observed that the higher the Er³⁺ (Yb³⁺) concentration, the higher the sample temperature.

Figure 10

Dependence of sample temperature on the 1550 nm laser irradiation time and working current for the sample with different Er³⁺ and Yb³⁺ concentrations.

Conclusion

YNbO₄ phosphors co-doped with various concentrations of Er³⁺ and Yb³⁺ were synthesized via a high temperature solid-state reaction method. XRD results indicated that monoclinic YNbO₄ phosphors were obtained. The concentration dependent UC luminescence properties were analyzed and it was found that the UC luminescence intensities greatly depended on the doping concentrations of Er³⁺ and Yb³⁺. By the analysis of the laser working current dependent UC luminescence spectra, it was confirmed that three-photon processes were dominated in the red UC emissions, and three-photon and four-photon processes co-existed in the green UC emissions under 1550 nm excitation. In addition, the study of the temperature sensing properties showed that the temperature sensing sensitivity is concentration dependent, and the lower concentrations of Er³⁺ and Yb³⁺, the higher temperature sensitivity. Furthermore, the YNbO₄: Er³⁺/Yb³⁺ phosphors can quickly reach thermal equilibrium excited by 1550 nm laser with high laser power density, and the sample temperature induced by 1550 nm laser irradiation was significantly dependent on both the RE doping concentration and the laser working current. The above results demonstrate that the concentrations of Er³⁺ and Yb³⁺ have significant influences on the UC luminescence processes, the temperature sensitivity of Er³⁺ and the laser-induced thermal effect of YNbO₄: Er³⁺/Yb³⁺ phosphors. Our results can provide a good reference for choosing proper doping concentrations of RE ions in the study of Er³⁺ related UC luminescence and temperature sensing.