Improving Optical Temperature Sensing Performance of Er³⁺ Doped Y₂O₃ Microtubes via Co-doping and Controlling Excitation Power

Abstract

This work presents a new method to effectively improve the optical temperature behavior of Er³⁺ doped Y₂O₃ microtubes by co-doping of Tm³⁺ or Ho³⁺ ion and controlling excitation power. The influence of Tm³⁺ or Ho³⁺ ion on optical temperature behavior of Y₂O₃:Er³⁺ microtubes is investigated by analyzing the temperature and excitation power dependent emission spectra, thermal quenching ratios, fluorescence intensity ratios, and sensitivity. It is found that the thermal quenching of Y₂O₃:Er³⁺ microtubes is inhibited by co-doping with Tm³⁺ or Ho³⁺ ion, moreover the maximum sensitivity value based on the thermal coupled ⁴S_3/2/²H_11/2 levels is enhanced greatly and shifts to the high temperature range, while the maximum sensitivity based on ⁴F_9/2(1)/⁴F_9/2(2) levels shifts to the low temperature range and greatly increases. The sensitivity values are dependent on the excitation power, and reach two maximum values of 0.0529/K at 24 K and 0.0057/K at 457 K for the Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ at 121 mW/mm² excitation power, which makes optical temperature measurement in wide temperature range possible. The mechanism of changing the sensitivity upon different excitation densities is discussed.

Introduction

Recently, optical temperature sensing behavior based on up-conversion luminescence of rare earth ion-doped phosphors have received much more attention since they can provide a non-contact temperature measurement in nanometer and submicron scale through probing the temperature-dependent fluorescence intensity ratio (FIR) of two adjacent thermally coupled energy levels^{1,2,3,4,5,6,7,8,9,10,11}. The non-contact FIR technique is superior to the conventional temperature measurements, shows large spatial resolution and high accuracy of detection¹². At present, trivalent rare earth ions, such as Er³⁺, Ho³⁺, Tm³⁺, Eu³⁺, and Pr³⁺, have been used as activators to study optical temperature sensing behaviors^{13,14,15,16,17,18}. Among these ions, Er³⁺ is preferred for temperature sensing due to the large energy gap (about 800 cm⁻¹) and small overlap of the two emission peaks from ²H_11/2 and ⁴S_3/2 levels^19,20,21,22. So the optical thermometry based on up-conversion emission of Er³⁺ doped phosphors has been explored in different hosts by using infrared excitation sources^{12,13,14,15,16,17,18}. These works reported that the optical temperature sensitivity of Er³⁺ doped phosphors was mainly dependent on host types, and lacked the investigation on the role of excitation powers and doping concentrations on the optical temperature behaviors.

However, Marciniak and Bednarkiewicz’s group observed that the highest sensitivity was reached 2.88%/K for LiYbP₄O₁₂:0.1%Er³⁺ nanocrystals upon pulsed excitation at average power below 25 mW/cm², while the same material displayed lower ~0.5%/K sensitivity at higher 50–300 mW/cm² excitation intensities²³. Prasad’s group observed that the intensity ratio of green and red emissions were dependent on the excitation power density²⁴. Li reported that the temperature sensing property of Er³⁺-Yb³⁺ co-doped NaGdTiO₄ was dependent on the Er³⁺ concentration²⁵. It was observed that the sensitivity values were strongly dependent on the excitation powers and excitation modes^{2, 26}. Thus, it is interesting to systematically study the concentration and excitation powers dependent optical temperature behaviors in different materials.

Phosphors with high thermal stability are excellent candidate materials to be used for optical temperature sensing. Compared with the fluorides, Y₂O₃ has the advantages of a high melting point, wide bandgap, high solubility between Y³⁺ and Er³⁺, and good transparency in ultraviolet and infrared range²⁷. Based on the green emissions from thermally coupled energy levels of ²H_11/2 and ⁴S_3/2, the optical temperature sensing was studied in Er³⁺-Yb³⁺ co-doped Y₂O₃ bulks and nano-materials^{27,28,29,30,31,32}. However, the aforementioned works have not dealt with excitation powers dependent optical temperature sensitivity, and lack to explore how to improve the optical temperature behaviors. The optical temperature property of Er³⁺ doped phosphors was determined by the population of the thermally coupled energy levels of ²H_11/2 and ⁴S_3/2 ¹². If the emission intensity ratios from the ²H_11/2 and ⁴S_3/2 is adjusted, the temperature sensitivity will change greatly. As reported that the energy gaps between ³F₄ and ³H₆ of Tm³⁺ ion, as well as between ⁵I₇ and ⁵I₈ of Ho³⁺ ion, were equal to the energy gap between ²H_11/2 and ⁴F_9/2 of Er³⁺ ion¹⁹. Thus, the population of ²H_11/2 and ⁴S_3/2 levels of Er³⁺ can be adjusted through the cross relaxation energy transfer between Er³⁺ and Tm³⁺ (or Ho³⁺). In this case the optical temperature sensitivity will be easy controlled through controlling the concentration of the Tm³⁺ or Ho³⁺ ion. Considering the structure of energy levels of Tm³⁺ and Ho³⁺, in this work, we propose a method to improve the optical temperature behavior of Er³⁺ doped Y₂O₃ through adjusting the green emission ratios with the cross-relaxation energy transfer between Er³⁺ and Tm³⁺ (or Ho³⁺). The Er³⁺ doped, Er³⁺-Tm³⁺ co-doped, and Er³⁺-Ho³⁺ co-doped Y₂O₃ microtubes are synthesized, and their optical temperature behaviors are studied by controlling the excitation power of the laser and the concentration of Tm³⁺ or Ho³⁺ ions. It is observed that the optical temperature sensitivity of Er³⁺-doped Y₂O₃ microtubes is significantly improved compared to the optical temperature sensitivity of Er³⁺-Yb³⁺ co-doped Y₂O₃ bulks and nano-materials.

Results

The low and high-magnification SEM images of Y₂O₃:1%Er³⁺, 1.5%Tm³⁺ in Fig. 1(a) and (b) show that the feature of the sample is a hollow and tubular column, and the average diameter is 0.7μm. The EDS spectrum in Fig. 1(c) shows that the samples consist of O, Er, Tm, and Y, which is in good accordance with the initial elements in precursor solution. All the synthesized Y₂O₃:1%Er³⁺ and Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ microtubes were analyzed also by the SEM and EDS. It is found that the shapes of all the samples are similar, and their sizes have no obvious change. Figure 2 displays the powders XRD patterns of Y₂O₃:1%Er³⁺, 1.5%Tm³⁺ and Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ microtubes. The position and relative intensity of all the diffraction peaks can be readily indexed to the pure cubic Y₂O₃ according to the JCPDS file no. 71-0099. No peak is recorded from other phases or impurities, which indicates the phase of the sample was pure Y₂O₃ tubes. From XRD patterns, one can find that the Y₂O₃ microtubes grow along the single direction (222) plane.

Figure 1

(a) SEM image, (b) enlarged SEM image, and (c) EDS spectrum of Y₂O₃:1%Er³⁺, 1.5%Tm³⁺.

Figure 2

Powder XRD patterns of (a) Y₂O₃:1%Er³⁺, 1.5%Tm³⁺ and (b) Y₂O₃:1%Er³⁺, 0.5%Ho³⁺. Below is the standard card.

In order to study the role of Tm³⁺ and Ho³⁺ ions, the up-conversion spectra, red to green intensity ratios and CIE (X, Y) chromaticity coordinates of Y₂O₃:1%Er³⁺ co-doped Tm³⁺ and Ho³⁺ are given in Fig. 3. Figure 3(a) shows the seven emission bands centered at 524 nm, 537 nm, 552 nm and 660 nm, 680 nm, 812 nm, 843 nm, which corresponds to the ²H_11/2 → ⁴I_15/2 (524–537 nm), ⁴S_3/2 → ⁴I_15/2 (552 nm), ⁴F_9/2 → ⁴I_15/2 (660–680 nm), and ⁴I_9/2 → ⁴I_15/2 (843 nm) transitions of Er³⁺ ions and ³H₄ → ³H₆ (812 nm) transition of Tm³⁺ ions. It is observed that the luminescence intensity of the green and 840 nm infrared emission decrease with increasing Tm³⁺ concentration, and the intensity of red emission first increases and then decreases. The intensity decrease of the green and red emissions is attributed to the quenching induced by the cross relaxation energy transfer between Tm³⁺ and Er³⁺: ⁴F_7/2 + ³H₆ → ⁴I_9/2 + ³H₅ ³¹. Figure 3(b) shows the red-green intensity ratio increases with the increase of Tm³⁺ concentration, and luminescent color is tunable from green to red with the increase of Tm³⁺ concentration. Similar spectrum modulation and color adjustment can be achieved through co-doping Ho³⁺ into Y₂O₃:1%Er³⁺ microtubes, as shown in Fig. 3(c) and (d). The intensities of the green and red emissions show the irregularly change with increasing Ho³⁺ concentration, while the intensity of infrared emission has no obvious change. The intensity increase of the green and red emissions of Y₂O₃:1%Er³⁺, when the Ho³⁺ concentration is more than 1%, is attributed to the contribution of Ho³⁺ on green and red emissions through the ⁵S₂ → ⁵I₈ (green) and ⁵F₅ → ⁵I₈ (red) transitions³³. The Ho³⁺ doping induces the increase of the red-green intensity ratio, and the enhancement of the yellow color, as shown in Fig. 3(d). After doping with Tm³⁺ and Ho³⁺, the emission intensity, the red-green intensity ratio, and the color of Y₂O₃:Er³⁺ are adjusted efficiently. This means that it is possible to adjust the optical temperature behavior of Y₂O₃:Er³⁺ at a high temperature through doping Tm³⁺ and Ho³⁺ ions.

Figure 3

(a) Emission spectra, (b) red to green intensity ratio and CIE (X, Y) chromaticity coordinates diagram of 1%Er³⁺, x%Tm³⁺ co-doped Y₂O₃ (x = 0, 0.2, 0.5, 1, 1.5). (c) Emission spectra and (d) Red to green intensity ratio and CIE (X, Y) chromaticity coordinates diagram of 1%Er³⁺, x%Ho³⁺ co-doped Y₂O₃ (x = 0, 0.2, 0.5, 1, 1.5).

Temperature-dependent emission spectra of Y₂O₃:1%Er³⁺ is shown in Fig. 4(a). It can be seen that the red and green emissions continuously decrease by increasing the temperature from 298 K to 573 K, without changing the peak positions of the emissions. The color shifting from yellow to green in Fig. 4(b) indicates the inhomogeneous decrease of green and red emissions induced by the high temperature. It is necessary to study the dependence of the fluorescence intensity ratio (R) on the temperature for the adjacent emission bands. The relation between R and T is expressed as:

$$LnR=-a/T+b$$

(1)

where a is constant, b is a correction term for the comprehensive population of thermally coupled energy levels induced by the thermal population, nonradiative relaxation and so on^{2, 34}. Relative sensitivity is one of the key parameters to determine the suitability for optical thermometry, and is defined as

$$S=\frac{dR}{dT}=\frac{a}{{T}^{2}}{e}^{\frac{bT-a}{T}}$$

(2)

where a and b are the constants from Eq. (1). Figure 4(c) shows temperature-dependent emission intensity ratios of several adjacent emission bands. The experimental data points can be fitted well with a line model. The slope values of fitted lines are dependent on the combination types of adjacent emission bands. It means that the sensitivity values are different when we use the different adjacent emission bands as the thermal coupled levels. Figure 4(d) shows the temperature dependent sensitivity values of five thermal coupled levels. One can find that all the sensitivity values increase and then decrease with the temperature increase, exhibiting the maxima values at different temperature points. The maximum values at (101 K, 0.0050/K), (461 K, 0.0027/K), (360 K, 0.0018/K), (413 K, 0.0044/K), (52 K, 0.0142/K) are observed for the adjacent thermal coupled levels of 524 nm/537 nm, 524 nm/552 nm, 537 nm/552 nm, (524 + 537) nm/552 nm, 660 nm/680 nm. Notably, the adjacent thermal coupled levels of (524 + 537) nm/552 nm shows the large intensity in the high temperature range. The adjacent thermal coupled levels of 660 nm/680 nm shows the very large intensity in the low temperature range. As reported, the Er³⁺-Yb³⁺ co-doped Y₂O₃ sphere nano-particles showed the maximum sensitivity value of 0.0044/K at 427 K²⁸, and the Er³⁺-Yb³⁺-Eu³⁺ tri-doped Y₂O₃ sphere nanoparticles showed the maximum sensitivity value of 0.0103/K at 593 K²⁷. In contrast, our Er³⁺- doped Y₂O₃ microtubes show a large sensitivity value of 0.0142/K in the low temperature range, and an excellent sensitivity value of 0.0044/K in the high temperature range. Most of the sensors based on up-conversion luminescence of Er³⁺ ion showed excellent sensitivity properties at the high temperature of more than 300 K^{12, 14}, while it was reported rarely on optical thermometry below room temperature.

Figure 4

Temperature dependent (a) emission spectra, (b) CIE (X, Y) chromaticity coordinates diagram, (c) Arrhenius plots of emission intensity ratios, and (d) sensitivity of Y₂O₃:1%Er³⁺at 184 mW/mm² excitation.

The influence of Tm³⁺ and Ho³⁺ on the optical temperature behaviors of Y₂O₃:1%Er³⁺ is studied through co-doping 0.2 mol% Tm³⁺ and 0.5 mol% Ho³⁺ into Y₂O₃:1%Er³⁺, as shown in Figs 5 and 6. Compared with the Fig. 4, it is evident that the emission spectra, CIE chromaticity coordinates, and R values of Y₂O₃:1%Er³⁺ are adjusted after co-doping with Tm³⁺ or Ho³⁺. Importantly, after co-doping with Tm³⁺ and Ho³⁺ ions, the maximum sensitivity value based on the thermal coupled levels of (524 + 537) nm/552 nm, is greatly enhanced and shifts to the higher temperature range, while the maximum sensitivity value based on the thermal coupled levels of 660 nm/680 nm, shifts to lower temperature range and increases a lot. This makes it possible to achieve the optical temperature measurement in the low temperature range.

Figure 5

Temperature dependent (a) emission spectra, (b) CIE (X, Y) chromaticity coordinates diagram, (c) Arrhenius plots of emission intensity ratios, and (d) sensitivity of Y₂O₃:1%Er³⁺, 0.2%Tm³⁺ at 184 mW/mm² excitation.

Figure 6

Temperature dependent (a) emission spectra, (b) CIE (X, Y) chromaticity coordinates diagram, and (c) Arrhenius plots of emission intensity ratios, and (d) sensitivity of Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ at 184 mW/mm² excitation.

It is necessary to study the thermal stability of thermal coupled levels in the process of optical temperature sensing. The thermal stability of emission bands can be determined by the number change of photons involved in the up-conversion processes at a different temperature. The up-conversion emission intensity I and excitation power P is expressed as follows:

$$I\propto {P}^{n}$$

(3)

where I is the emission intensity, P is incident pump power, and n is the number of pump photons absorbed in the up-conversion process³⁵. Figure 7 shows the double logarithmic plots of the emission intensity I as a function of pump power P of the Y₂O₃:Er³⁺. The fit results indicate that three infrared photons are needed to emit green and red luminescence at 298 K and 573 K. After doping Tm³⁺ and Ho³⁺ into Y₂O₃:Er³⁺, the fit results indicate that two infrared photons are needed to emit green and red luminescence at 298 K and 573 K, as shown in Figs 8 and 9. The decrease of slope values for green and red emissions means that the up-conversion process becomes easy to occur at the same excitation power. Thus, the thermal stability of Y₂O₃:Er³⁺ microtubes is improved through co-doping Tm³⁺ and Ho³⁺ ions.

Figure 7

Log-log plots of emission intensity and pumping power for Y₂O₃:1%Er³⁺ at (a) 298 K and (b) 573 K.

Figure 8

Log-log plots of emission intensity and pumping power for Y₂O₃:1%Er³⁺, 0.2%Tm³⁺ at (a) 298 K and (b) 573 K.

Figure 9

Log-log plots of emission intensity and pumping power for Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ at (a) 298 K and (b) 573 K.

M. Pollnau observed that the population processes of green and red emissions were adjusted by the excitation power density³⁵. It is necessary to study the excitation power-dependent optical temperature behaviors of Y₂O₃:1%Er³⁺, Y₂O₃:1%Er³⁺, 0.2%Tm³⁺, and Y₂O₃:1%Er³⁺, 0.5%Ho³⁺. To evaluate the influence of excitation powers on luminescence quenching, the thermal quenching ratio (TRQ) of Er³⁺ emission bands at different excitation powers are studied in Figs 10, 11 and 12. The thermal quenching ratio (TRQ) of emission bands induced by the temperature change is defined as follows

$$TRQ=1-\frac{{I}_{T}}{{I}_{R}}$$

(4)

where I _T is luminescence intensity at a different temperature T, and I _R is luminescence intensity at room temperature³⁶. From Fig. 10(a) and (b), it is obvious that the TRQ of green and red emissions of Y₂O₃:1%Er³⁺ is strongly dependent on temperature and excitation powers. The values of TRQ of green and red emissions increase with the increase of temperature, change irregularly with the increase of excitation powers. The red-to-green intensity ratio is dependent on both temperature and excitation powers, and shows the large values at the excitation power of 121 mW/mm², as shown in Fig. 10(c). After co-doping with the Tm³⁺ and Ho³⁺, the TRQ values of green and red emissions and red-to-green emission intensity ratios of Y₂O₃: 1%Er³⁺ are adjusted, as shown in Figs 11 and 12.

Figure 10

Excitation powers dependent TRQ of (a) green and (b) red emissions of Y₂O₃:1%Er³⁺, and (c) red to green intensity ratio of Y₂O₃:1%Er³⁺.

Figure 11

Excitation powers dependent TRQ of (a) green and (b) red emissions of Y₂O₃:1%Er³⁺, 0.2%Tm³⁺, and (c) red to green intensity ratio of Y₂O₃:1%Er³⁺, 0.2%Tm³⁺.

Figure 12

Excitation powers dependent TRQ of (a) green and (b) red emissions of Y₂O₃:1%Er³⁺, 0.5%Ho³⁺, and (c) red to green intensity ratio of Y₂O₃:1%Er³⁺, 0.5%Ho³⁺.

The influence of excitation powers on the sensitivity values for the Y₂O₃:1%Er³⁺ are studied in Fig. 13. At different excitation powers, based on the ratio of (524 + 537) nm/552 nm, the maximum sensitivity value of 0.0049/K at 471 K is achieved at 30 mW/mm², as shown in Fig. 13(a). Based on the ratio of 660 nm/680 nm, the maximum sensitivity value of 0.0382/K at 34 K is achieved at 322 mW/mm², as shown in Fig. 13(b). After co-doping with Tm³⁺, the maximum sensitivity values are significantly enhanced and shift to the high temperature range with various Tm³⁺ concentrations, as shown in Fig. 14(a) and (b). The sensitivity value based on the ratio of (524 + 537) nm/552 nm achieves the maximum value at (459 K, 0.0055/K) when the Tm³⁺ concentration is 0.2 mol%. The sensitivity value based on the ratio of 660 nm/680 nm achieve the maximum value at (37 K, 0.0235/K) when the Tm³⁺ concentration is 1.0 mol%. Furthermore, the excitation power dependent sensitivity is studied in Fig. 14(c) and (d). The maximum sensitivity values change irregularly with the increase of the excitation powers, and reach the maximum value of (504 K, 0.0056/K) at 30 mW/mm² for the ratio of (524 + 537) nm/552 nm, and the maximum value of (22 K, 0.0282/K) at 85 mW/mm² for the ratio of 660 nm/680 nm. Similarly, the influence of Ho³⁺ concentration and excitation powers on the sensitivity values are also studied in Fig. 15. The optimized Ho³⁺ concentration is obtained is 0.5 mol%. For the Y₂O₃:1%Er³⁺, 0.5%Ho³⁺, the maximum sensitivity values change irregularly with the increase of the excitation powers, and reach the maximum value of (457 K, 0.0057/K) at 184 mW/mm² for the ratio of (524 + 537) nm/552 nm, and the maximum value of (24 K, 0.0529/K) at 121 mW/mm² for the ratio of 660 nm/680 nm. Compared with the Y₂O₃:1%Er³⁺, the maximum sensitivity value increases from 0.0049/K (0.0382/K) to 0.0057/K (0.0529/K) through co-doping with the 0.5 mol% Ho³⁺ ions. In order to compare the sensitivity values of Y₂O₃ doped with other ions for the fluorescence thermometric study, the reported sensitivity values based on the up-conversion fluorescence of Er³⁺ are listed in Table S1 in the supplementary information. It can be seen that the sensitivity values in this paper are higher than that of other Y₂O₃ materials not only in the high temperature range but also in the low temperature range. It means that it is a good method to improve the sensitivity of Y₂O₃:Er³⁺ through co-doping Ho³⁺ ions.

Figure 13

Excitation powers dependent sensitivity of Y₂O₃:1%Er³⁺.

Figure 14

Doped concentration dependent sensitivity of Y₂O₃:1%Er³⁺, x%Tm³⁺ (x = 0, 0.2, 0.5, 1.0, 1.5), and excitation powers dependent sensitivity of Y₂O₃:1%Er³⁺ 0.2%Tm³⁺.

Figure 15

Doped concentration dependent sensitivity of Y₂O₃:1%Er³⁺, y%Ho³⁺ (x = 0, 0.2, 0.5, 1.0, 1.5, 2.0), and excitation powers dependent sensitivity of Y₂O₃:1%Er³⁺, 0.5%Ho³⁺.

The influence of doping concentration and excitation powers on the sensitivity values for the Y₂O₃:1%Er³⁺ are also studied at the fixed temperatures, as shown in Fig. 16. One can find that the sensitivity values from the ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels increase and then decrease with the increase of doping concentrations of Tm³⁺ and Ho³⁺ ions, while the sensitivity values from the ²H_11/2/⁴S_3/2 thermal coupled levels change irregularly with the increase of doping concentrations, as shown in Fig. 16(a) and (b). The optimized Tm³⁺ and Ho³⁺ concentrations are 0.2 mol% and 0.5 mol%, respectively. The sensitivity values from the ²H_11/2/⁴S_3/2 and ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels show several oscillating curves with the increase of excitation powers, as shown in Fig. 16(c) and (d). For the Y₂O₃:1%Er³⁺, 0.2%Tm³⁺, the sensitivity value based on ²H_11/2/⁴S_3/2 thermal coupled levels reaches the maximum value at 30 mW/mm², and the sensitivity value based on ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels reaches the maximum value at 85 mW/mm², as shown in Fig. 16(c). For the Y₂O₃:1%Er³⁺, 0.5%Ho³⁺, the sensitivity value based on ²H_11/2/⁴S_3/2 thermal coupled levels reaches the maximum value at 184 mW/mm², and the sensitivity value based on ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels reaches the maximum value at 121 mW/mm², as shown in Fig. 16(d).

Figure 16

Doping concentration and excitation powers dependent sensitivity of Y₂O₃:1%Er³⁺ 0.2%Tm³⁺ and Y₂O₃:1%Er³⁺, 0.5%Ho³⁺ samples.

To study the influence of doping concentration and excitation powers on the optical temperature behaviors, the dynamic balance rate-equation models for the energy transfer between Er³⁺ and Tm³⁺(or Ho³⁺) are established in Figs S1 and S2 in the supplementary information. The population dynamic process of excited states is simulated by using the eight-level model. The population density of the ²H_11/2/⁴S_3/2 and ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels can be obtained in the Equations S(8), S(9), S(17), and S(18). It is obvious that the values of N ₃ and N ₄ are dependent on not only the excitation power (ρ) and doping concentration (N₀, N₆), but also the cross relaxation rates (W _c1, W _c2, W _c3, and W _c4), nonradiative decay rates (W _ij ), and radiative transition rates (A _ij ). Notably, the cross relaxation rates, nonradiative decay rates, and radiative transition rates are strongly dependent on the temperature³⁷. It means that the population processes of the ²H_11/2/⁴S_3/2 and ⁴F_9/2(1)/⁴F_9/2(2) thermal coupled levels are determined by the excitation power, doping concentration and temperature. It is a complex multi-field coupling effect. Thus, the sensitivity shows irregularly change with the increase of excitation powers and doping concentrations.

Briefly speaking, at low excitation power, the thermal coupled levels of ⁴S_3/2 and ²H_11/2 are populated by the ground state absorption (GSA) and excited state absorption (ESA), and then the multiphonon nonradiative relaxation (NR) from ⁴F_7/2 level, shown in Fig. S1. The thermal coupled levels of ⁴F_9/2(1) and ⁴F_9/2(2) are populated by the NR process from the ⁴S_3/2 level. According to the Boltzmann distribution^{14, 15}, the population of the ²H_11/2 level increases with respect to the ⁴S_3/2 with the increase of the temperature, owe to the low energy difference between ⁴S_3/2 and ²H_11/2 levels. The thermalization of the ²H_11/2 level is dominant and the depopulation induced by the NR process can be neglected. At high excitation power, the population saturation effect of the ⁴S_3/2 and ²H_11/2 levels can be observed^{35, 38}. Therefore, the population of ²H_11/2 level changes a little with the temperature. The NR is easy to occur in the case of small energy difference between adjacent energy levels at the high temperature^{19, 20}. The NR process is dominant to depopulate the ²H_11/2 level, due to the small energy difference between ²H_11/2 and ⁴S_3/2 (ΔE = 968 cm⁻¹). Thus, the R in equation 1 decreases with the increase of excitation power density, due to the fact that the NR possibility increases with the temperature increase. As a result, the sensitivity decreases at high excitation power density. Importantly, after co-doping Tm³⁺ and Ho³⁺ ion, the sensitivity values shift to higher and lower temperature ranges, and are greatly increased. It is attributed to the fact that the populations of ⁴S_3/2/²H_11/2 and ⁴F_9/2(1)/⁴F_9/2(2) levels are adjusted by the cross relaxation process, such as CR1 and CR3. With the concentration further increase of the Tm³⁺ and Ho³⁺ ions, the green and red emissions are quenched, due to the cross relaxation process, such as CR2 and CR4.

Conclusions

In summary, we explore a method to improve the photoluminescence and optical temperature sensing of Er³⁺ doped Y₂O₃ microtubes through combining the ion doping with the control of excitation powers. It is found that the optical temperature behaviors of Er³⁺ doped Y₂O₃ microtubes are strongly dependent on the ion doping and excitation powers. After doping Tm³⁺ or Ho³⁺ ions, the spectrum of Er³⁺ doped Y₂O₃ microtubes is modified, the thermal quenching behavior of Y₂O₃:Er³⁺ microtubes is inhibited, and the optical temperature sensitivity is significantly enhanced. It is achieved that the maximum sensitivity value is 0.0529/K at low temperature range of less than 250 K while it is 0.0057/K at more than 250 K by controlling the excitation power to 121 mW/mm². The maximum sensitivity value of 0.0529/K at 24 K is superior to that of our earlier report. It makes up the lack of optical temperature detection in the low temperature range below 250 K.

Methods

All starting materials are Y₂O₃ (99.99%), Er₂O₃ (99.99%), Tm₂O₃ (99.99%), Ho₂O₃ (99.99%), hydrochloric acid (AR), NaOH (AR), ethanol (AR). All chemicals were used without further treatment, and deionized water was used for all experiments.

The Re₂O₃ (Re = Er, Y, Tm, and Ho) was dissolved in hydrochloric acid, and then the solution was heated to evaporate the water completely. The obtained rare earth metal trichloride (ReCl₃) was dissolved in deionized water to prepare the solutions of ReCl₃ (0.2 mol L⁻¹). Er³⁺ doped Y₂O₃ microtubes were prepared by a hydrothermal method. In a representative synthesis process, an aqueous solution of 9.90 mL YCl₃ (0.2 mol L⁻¹), 0.10 mL ErCl₃ (0.2 mol L⁻¹) was mixed with 28 mL of distilled water under thorough stirring. It was then vigorously stirred by a magnetic stirrer at room temperature for 30 min, while 3.5 mL of a 5 mol L⁻¹ NaOH solution was slowly added in drops. The colloidal solution was then transferred into an Teflon vessel at 473 K for 24 h. The final products were collected, washed several times with ethanol, and purified by centrifugation. Samples were then dried in an oven for 6 h at 373 K. Finally, the dried powders were sintered at 1173 K for 3 h in an electric annealing furnace. After then, the samples were cooled down rapidly to room temperature. The same method was used for Er³⁺/Tm³⁺ and Er³⁺/Ho³⁺ co-doped Y₂O₃ microtubes.

The structure of the sample was investigated by X-ray diffraction (XRD) using X’TRA (Switzerland ARL) equipment provided with Cu tube with Kα radiation at 1.54056 Å. The size and shape of the sample were observed by a JSM-IT300 scanning electron microscope (SEM) (JEOL Ltd., Tokyo, Japan) equipped with an energy dispersive X-ray spectrometer (EDS). Luminescence spectra were obtained by the Acton SpectraPro Sp-2300 Spectrophotometer with a photomultiplier tube equipped with a 980 nm laser as the excitation source. Different temperature spectra were obtained in the range 298–573 K by using an INTEC HCS302 Hot and Cold System.