Optical Temperature-Sensing Performance of La₂Ce₂O₇:Ho³⁺ Yb³⁺ Powders

Abstract

In this paper, La₂Ce₂O₇ powders co-activated by Ho³⁺ and Yb³⁺ were synthesized by a high temperature solid-state reaction. Both Ho³⁺ and Yb³⁺ substitute the La³⁺ sites in the La₂Ce₂O₇ lattice, where the Ho³⁺ concentration is 0.5 at.% and the Yb³⁺ concentration varies in the range of 10~18% at.%. Pumped by a 980 nm laser, the up-conversion (UC) green emission peak at 547 nm and the red emission at 661 nm were detected. When the doping concentration of Ho³⁺ and Yb³⁺ are 0.5 at.% and 14% at.%, respectively, the UC emission reaches the strongest intensity. The temperature-sensing performance of La₂Ce₂O₇:Ho³⁺ with Yb³⁺ was studied in the temperature range of 303–483 K, where the highest relative sensitivity (S_r) is 0.0129 K⁻¹ at 483 K. The results show that the powder La₂Ce₂O₇:Ho³⁺, Yb³⁺ can be a potential candidate for remote temperature sensors.

1. Introduction

Since the concept of UC luminescence was first proposed by the French scientist Auzel in 1966 [1], it has been widely concerned in the application fields of remote thermometers [2], stimulated emission depletion [3], solar cells [4], etc. Remote thermometers based on UC luminescent materials are suitable for harsh environments, such as strong magnetic fields and corrosion [5]. Theoretically, the fluorescence intensity, fluorescence intensity ratio (FIR), fluorescence lifetime, and full width at half maximum (FWHM) of luminescent materials will change with the variation in temperature [6]. Based on these characteristics, the purpose of temperature sensing can be feasibly realized.

The optical properties of fluorescent materials are usually temperature-dependent within a certain temperature range. However, it should be pointed out that the measurement of fluorescence lifetime and intensity is highly dependent on the intensity of excitation light source, background noise, and other factors, so it is difficult to ensure the accuracy of the temperature measurement based on the above factors. At present, most of the research works on optical temperature measurement are based on FIR with good anti-interference ability [7,8,9,10,11]. Moreover, the UC luminescent materials activated by rare earth ions usually have abundant but narrow linear luminescence bands, which naturally feature more excellent FIR, and make it possess unique advantages in the field of temperature sensing.

UC luminescent materials for optical temperature measurement need to have high luminescence intensity, for which the selection of the host materials is very important. In recent years, the research on A₂B₂O₇ compounds has been blooming, a general formula that describes most pyrochlores as A₂B₂O₆O’, where the A site is occupied by a 3+ or 2+ cation with eight-fold coordination, the B site is occupied by a 4+ or 5+ cation with six-fold coordination, and there are two distinct oxygen atom sites [12]. At present, the reported A₂B₂O₇ compounds are mainly La₂Ce₂O₇ [13], Gd₂Ce₂O₇ [14], Nd₂Ce₂O₇ [15], etc., and the main application scenario focuses on the application of a thermal barrier coating, but few reports are found in the field of temperature sensing based on the fluorescence performance. La₂Ce₂O₇ belongs to the structure of defective fluorite, where the A site is occupied by La³⁺, and the B site is occupied by Ce⁴⁺, as shown in Figure 1a. La₂Ce₂O₇ has a high melting point (>2000 °C), excellent chemical stability, and high temperature phase stability and catalytic performance, which is widely used in thermal barrier coatings, infrared radiation ceramic materials, catalytic materials, etc. [16,17]. However, reports on the optical temperature-sensing performance of rare earth-doped La₂Ce₂O₇ materials are few [18,19].

Due to the small energy gap between ²H_11/2 and ⁴S_3/2 of the Er³⁺ ions, the FIR conforms to the Boltzmann distribution and has a high sensitivity to temperature variation; Er³⁺ has been most extensively investigated by researchers [19,20]. In addition, in recent years, there have been numerous studies on the temperature-sensing performance of Ho³⁺-activated phosphor materials [21,22]. Particularly, some studies have shown that Ho³⁺-doped UC luminescent materials have excellent temperature-sensing performance [23,24]. Ho³⁺ can efficiently emit colorful UC fluorescence with the assistance of a Yb³⁺ sensitizer under an excitation of the 980 nm near-infrared laser [25,26]. Govind B. Nair et al. [27] prepared Ho³⁺/Yb³⁺ co-doped BaY₂F₈, and investigated the UC luminescence with temperature change in the temperature range of 303–623 K, and obtained the maximum relative sensitivity (S_r) of 0.006051 K⁻¹ at 303 K.

In this work, a series of La₂Ce₂O₇:Ho³⁺ Yb³⁺ powders were prepared by high temperature solid-state reaction. The temperature dependence of the UC emission intensity has been investigated. The results show that La₂Ce₂O₇:Ho³⁺ Yb³⁺ powders are good candidates for optical temperature sensing.

2. Experimental

2.1. Materials Preparation

High purity powder Ho₂O₃ (99.99%), Yb₂O₃ (99.99%), CeO₂ (99.99%), and La₂O₃ (99.999%) were used as raw materials and weighed according to the chemical composition of (Ho_0.005Yb_xLa_0.995−x)₂Ce₂O₇ (x = 0.10, 0.12, 0.14, 0.16, 0.18). Since La₂O₃ is easy to absorb moisture, it is kept in the oven at 120 °C before use. The mixed powders were ground with an agate grinder for 1 h. The resultant powder mixtures were thermally treated in a Muffle furnace by the following sintering procedure: raised to 1370 K from room temperature within 150 min, then raised to 1890 K within 120 min and kept for 300 min, and finally cooled down to room temperature naturally. For the following characterizations, the obtained products were ground into fine powders with an agate grinder for 1 h.

2.2. Characterizations

X-ray diffraction (XRD, Rigaku SmartLab, Japan, Akishima City, Tokyo) was used to analyze the phase and crystal structure of the (Ho_0.005Yb_xLa_0.995−x)₂Ce₂O₇ (x = 0.10, 0.12, 0.14, 0.16, and 0.18) samples. The scanning range is 20–90°, the step size is 0.02°, and the scanning speed is 8°/min. The UC emission spectra under excitation of a 980 nm laser were recorded by a fluorescence spectrometer (FLS1000, Edinburg Instruments, UK, Edinburgh) with a resolution of 0.05 nm. In addition, the temperature-dependent PL spectra were measured in the 303–483 K temperature range, and the thermal recovery property and repeatability of the samples were also measured on the fluorescence spectrometer.

3. Results and Discussion

Figure 1b shows the XRD θ-2θ scanning patterns of La₂Ce₂O₇ and (Ho_0.005Yb_xLa_0.995−x)₂Ce₂O₇ (x = 0.10, 0.12, 0.14, 0.16, and 0.18) powder samples thermally treated at 1890 K for 5 h. The diffraction patterns of all the Ho³⁺ and Yb³⁺ co-doped samples were the same as La₂Ce₂O₇, without other phases detected. The difference is that the diffraction peaks are all shifted to higher angles, indicating that Ho and Yb have been successfully doped into the lattice of La₂Ce₂O₇. The shift of the diffraction peaks to a higher angle is caused by lattice shrinkage by the substitution of La³⁺ (R_La³⁺ = 1.22 Å) ions with larger ionic radii by Ho³⁺ (R_Ho³⁺ = 1.03 Å) and Yb³⁺ (R_Yb³⁺ = 0.887 Å) ions with a smaller ionic radius.

Figure 2a shows the UC emission spectra of the (Ho_0.005Yb_xLa_0.995−x)₂Ce₂O₇ (x = 0.10, 0.12, 0.14, 0.16, and 0.18) powders under the excitation of a 980 nm laser (pumping power 150 mW) at room temperature. Although the doping amount of Yb³⁺ is various, the UC emission is mainly composed of a strong green emission peak at 547 nm (Ho³⁺: ⁵S₂/⁵F₄→⁵I₈) and a weak red emission peak at 661 nm (Ho³⁺: ⁵F₅→⁵I₈). When the substitution amount of Yb³⁺ reaches x = 0.14, both the UC intensity of the green and red reach the maximum value. And, when x > 0.14, the UC intensity was gradually decreased, which was caused by the energy loss in the energy transfer (ET) between the Yb³⁺ ions or during the Yb³⁺→Ho³⁺ ET process.

Figure 2b shows the UC emission intensity spectra of the (Ho_0.005Yb_0.14La_0.855)₂Ce₂O₇ sample under the excitation of a 980 nm laser with different pumping powers (200–2140 mW) at room temperature. It can be clearly seen that with the increase in laser power, the 547 nm UC emission intensity (I₅₄₇) and the 661 nm UC emission intensity (I₆₆₁) rapidly increases. The initial UC emission intensity increases with the increase in excitation power, then gradually decreases when the power reaches 1800 mW. When the excitation power is less than 1400 mW, the UC emission intensity changes rapidly, and when the power is more than 1400 mW, the UC emission intensity changes slowly. A hysteresis-like loop was observed for the Ho³⁺: ⁵F₄/⁵S₂→⁵I₈ transitions of ions when the pump power was increased from 200 mW to 2140 mW for the (Ho_0.005Yb_0.14La_0.855)₂Ce₂O₇ sample, and, conversely, for the excitation optical power down process. It can be seen from Figure 2b that with the pumping power increased, the UC emission intensity first increases and then decreases owing to the multi-phonon assisted non-radiative relaxations. However, in the “power-down” process, it does not take the same path, but rather follows a path with a slightly lower intensity and gives rise to a hysteresis curve. The same output power corresponds to two different emission intensities, and there is a loss in the intensity in the reverse process. The obtained hysteresis loop is clarifying the evidence of intrinsic optical bi-stability (IOB). The existence of emission intensity lag within a complete excitation power period is an indicator of inherent optical bi-stability [28]. At the same time, the increase in temperature caused by the increase in laser power is also one of the reasons for the hysteresis cycle of UC emission intensity.

According to the UC emission intensity theory, the UC emission intensity is proportional to the pumping power intensity, where n is the number of photons required for UC emission. In the UC emission process, the relationship between UC emission intensity (I) and excitation power (P) [29,30,31] can be expressed by the following formula:

$$I\propto P^n$$

(1)

The logarithm of pump power and the logarithm of luminescence intensity are used for linear fitting, and the slope of the line obtained is the number of photons required for UC emission, n, which can be obtained by the logarithm of Equation (1):

$$n=\frac{LnI}{LnP}$$

(2)

For the (Ho_0.005Yb_0.14La_0.855)₂Ce₂O₇ sample, the logarithmic relationship plotting between the intensity of the 547 nm and 661 nm UC emissions and the excitation power P is shown in Figure 2c. The slopes of the linear fitting for the 547 nm and 661 nm UC emissions are 1.23 and 1.15, respectively, indicating that both of the two UC emissions are based on the two-photon process. For the two-photon process, n should be equal to or approximately equal to 2, but the n value of both sets of experimental data are less than 2. This may be caused by the competition between linear decay and UC processes for the depletion in the intermediate excited states [32]. As reported by Xue et al. [33], when UC is dominant, the intensity is proportional to the excitation power density, and when the intermediate excited state energy level is dominated by linear decay, the intensity is proportional to the square of the excited power. Therefore, since n is closer to 1 (n = 1.23, 1.15), the linear decay of the intermediate excited state is dominant in this work.

In order to better understand the UC emission process of La₂Ce₂O₇:Ho³⁺ Yb³⁺, the energy level diagrams of Ho³⁺ and Yb³⁺ and the possible energy transfer during the UC emission process are provided in Figure 2d. Under the excitation of a 980 nm laser, the Yb³⁺:²F_7/2 ground state electrons absorb the energy and transit to the Yb³⁺:²F_5/2 excited state, while the Ho³⁺:⁵I₈ ground state electrons jump to the Ho³⁺:⁵I₆ excited state through the energy transfer (ET1) from Yb³⁺ to Ho³⁺. The intermediate level of Ho³⁺:⁵I₆ has a long life, which can continue to absorb energy and, in the meantime, transit to the excited state of Ho³⁺:⁵F_4/5/⁵S₂ of higher energy via excited state absorption (ESA). Some electrons decay from the Ho³⁺:⁵I₆ state to the Ho³⁺:⁵I₇ state through multi-phonon relaxation. Electrons in the Ho³⁺:⁵I₇ state can go up to the Ho³⁺:⁵F₅ state, which is also an ESA process. Finally, electrons transit from the Ho³⁺:⁵F_4/5/⁵S₂ and Ho³⁺:⁵F₅ states to the Ho³⁺:⁵I₈ ground state. Above all, the green and red UC emissions are generated.

In order to study the temperature-sensing performance of Yb³⁺ and Ho³⁺ co-doped La₂Ce₂O₇ UC-emitting powders, temperature-dependent UC emission spectra under the excitation of a 980 nm laser were measured in the range of 303–483 K, as shown in Figure 3a. The inset shows the enlarged view of the red UC emission spectra in the wavelength range of 640–680 nm. It can be observed that with the increase in temperature, the intensity of the green and the red UC emissions gradually weakens, due to the thermal quenching. With the increase in temperature, phonon vibration in the lattice is enhanced, leading to the increase in the probability of non-radiative relaxation, and the excessive energy is lost in the form of heat, resulting in the reduced UC emission intensity [10,34,35].

Specifically, as shown in Figure 3b, I₆₆₁ has no obvious tendency to change with temperature; however, I₅₄₇ decreased evidently when the temperature increased from 303 K to 483 K, which may be because the probability of non-radiative relaxation from ⁵F₄/⁵S₂ level to ⁵F₅ level increased with increasing temperature. At the same time, the increase in temperature will also produce thermal excitation to promote the electron transition to the higher energy level of ⁵F₄/⁵S₂. The different response of these two energy levels to temperature provides favorable conditions for obtaining excellent temperature sensitivity [36].

Non-thermally coupled energy levels involve two independent excited energy levels, each with a unique temperature dependence. FIR based on non-thermally coupled levels of ⁵F₄/⁵S₂→⁵I₈ and ⁵F₅→⁵I₈ can be used for optical temperature measurement. Since the intensity of 547 nm and 661 nm UC emissions are temperature-dependent, the FIR of corresponding energy levels can be used for temperature sensing. Figure 3c shows the plot of the FIR of the 661 nm and 547 nm UC emissions against the absolute temperature. The ratio I₆₆₁/I₅₄₇ increased from 0.13 to 0.46 with the temperature increasing from 303 K to 483 K. Figure 3d shows the plot of Ln (I₆₆₁/I₅₄₇) as a function of the inverse absolute temperature, which presents a straight line with a slope of −967.1158 and an intercept of 1.08179. In addition, the sensitivity is a very major parameter to estimate the performance of the sensor. The absolute sensitivity (S_a) and the relative sensitivity (S_r) can be given as the following [37]:

$$S_a=\left|\frac{dFIR}{dT}\right|$$

(3)

$$S_r=\left|\frac{1}{FIR}\frac{dFIR}{dT}\right|$$

(4)

Figure 4a,b show the variation in the S_a and S_r of samples with temperature within the temperature range of 303–483 K. It can be seen that the absolute sensitivity S_a of the sample increases with the increase in temperature. Under the excitation of a 980 nm laser, the maximal S_a value is 0.0059 K⁻¹ at 483 K. S_r shows fluctuating variation with the temperature, reaching a maximal value of 0.0129 K⁻¹ around 483 K.

Temperature resolution (δT) is an important parameter for evaluating the performance of temperature sensing, indicating the smallest detectable temperature change. The temperature resolution obtained through experiments is generally expressed as [38]

$$\delta T=\frac{1}{S_r}\frac{\delta FIR}{FIR}$$

(5)

In the above equation, δFIR represents the precision of the FIR, while δFIR/FIR is primarily determined by the properties of the detector. By conducting three measurements of the emitted spectrum and calculating their variances, the standard deviation of FIR at room temperature is determined to be 0.0378%. Utilizing this value, the temperature resolution can be computed, as illustrated in Figure 4c. It can be observed that the δT values range between 0.064 and 0.52, with an average δT value of around 0.25 K.

In order to investigate the thermal recovery performance of (Ho_0.005Yb_0.14La_0.855)₂Ce₂O₇ powder, five emission spectra were measured under 1400 mW: (a) when the temperature rises to 413 K, the emission spectrum was measured immediately; (b) when 413 K lasts for 5 min; (c) when 413 K lasts for 10 min; (d) when 413 K lasts for 20 min; and (e) when 413 K lasts for 40 min. As shown in Figure 4d, after 413 K lasts for 5 min, the UC emission intensity decreased significantly, but after holding the powder for 10 min, 20 min, and 40 min, the UC emission intensity was gradually increased and recovered to the height of “(a)”, indicating that the thermal recovery performance of La₂Ce₂O₇ powder is good.

The optical temperature-sensing performance of materials doped with Ho³⁺ and Yb³⁺ based on the FIR technology are shown in

Table 1. Optical temperature-sensing properties of Ho³⁺/Yb³⁺-doped materials.

Materials	Temperature Range (K)	λex (nm)	Sr (K−1) (Max)	References
NaLaMgWO6	293–553	980	0.01079 (508 K)	[2]
Ba0.77Ca0.23TiO3	93–300	980	0.0053 (93 K)	[10]
Tellurite glass	303–503	975	0.011 (503 K)	[24]
BaTiO3	303–513	980	0.0034 (303 K)	[25]
CaMoO4	303–543	980	0.0066 (353 K)	[26]
LiYF4	100–500	976	0.0129 (150 K)	[39]
Gd0.74Y0.2TaO4	330–660	980	0.0037 (660 K)	[40]
LaOF	298–548	980	0.00451 (298 K)	[41]
Y2O3	303–623	980	0.0064 (423 K)	[42]
(Ho0.005Yb0.14La0.855)2Ce2O7	293–428	980	0.0129 (428 K)	This work

. The relative sensitivity of (Ho_0.005Yb_0.14La_0.855)₂Ce₂O₇ phosphor is higher than that of most of the materials listed in Table 1, which shows that Ho³⁺ and Yb³⁺-doped La₂Ce₂O₇ powder is a potential candidate for optical temperature sensing.

4. Conclusions

In this work, Ho³⁺ and Yb³⁺ co-doped La₂Ce₂O₇ powders were prepared by solid-state reaction at 1890 K. Under excitation by a 980 nm laser, green UC emission peaks were observed at 547 nm and red UC emission peaks at 661 nm at room temperature, with the UC emission intensity decreasing as the temperature increased. A quite clear relationship can be observed between the Ho³⁺/Yb³⁺ UC emission and the temperature. The highest relative sensitivity was 0.0129 K⁻¹ at the temperature of 483 K. These properties indicate that La₂Ce₂O₇:Ho³⁺ Yb³⁺ powder is a potential candidate material for optical temperature sensing.