1. Introduction

In the past decade, optical temperature sensing based on the fluorescent response to temperature has attracted tremendous attention due to its outstanding merits, including contactless detection, quick response, and high spatial resolution [1–3]. Among the various approaches for optical thermometry, ratiometric method which uses temperature dependent fluorescence intensity ratio (FIR) as detecting parameter has been regarded as a promising technique benefiting from the immunity to some disturbances, for instance, fluorescence losses, unevenly distributed luminescent centers, and fluctuations of excitation intensity [4–8].

Recently, we reported a fluorescence intensity-based strategy taking advantage of temperature-induced shift of V-O charge transfer band (CTB) edge [9–10]. Drastic temperature dependence of the fluorescence was obtained under a certain excitation in the tail of the CTB caused by the shift of the CTB edge. However, such intensity-based method may suffer from some disturbances as we mentioned above, which may further influence the measurement accuracy. On the basis of this strategy, if another luminescence center whose fluorescence is immune to temperature or exhibits opposite temperature dependence is introduced as a reference, ratiometric method can be applied. Taking YVO₄:Eu³⁺ for an example, we found that Eu³⁺ emission increases with the increase of temperature ranging from 300 to 480 K under the 358 nm excitation which locates in the tail of the V-O CTB [11]. As for the additional luminescence center to be introduced serving as the reference, efficient f-f transition excitation should exist and be dominant around 358 nm. Otherwise, it may exhibit similar temperature dependence with Eu³⁺ emission.

Tb³⁺ may be a promising candidate for the reference due to the existence of several f-f transition excitations around the position of V-O CTB edge [12]. But, unfortunately, in YVO₄ lattice, both f-f transition excitations and CT state excitation are inefficient for Tb³⁺ emission due to the strong quenching effect [13]. However, in YPO₄ lattice which has the same crystal structure with YVO₄, Tb³⁺ can emit strong fluorescence under the characteristic f-f excitations [14–15]. Based on this, we conclude that partial substitution of V by P in YVO₄ lattice would be helpful to Tb³⁺ emission. Thus, both temperature-induced shift of the V-O CTB edge and Tb³⁺ emission originating from f-f transition excitation can be guaranteed. Therefore, we expect that the solid solution YV_1-xP_xO₄ may be an ideal matrix for temperature dependent FIR of Eu³⁺ to Tb³⁺.

To verify the feasibility of this proposal, rare earth ions doped YV_1-xP_xO₄ solid solutions were synthesized. Temperature dependent optical properties of the as-prepared samples were studied for ratiometric optical temperature sensing.

2. Experimental

Rare earth doped YV_1-xP_xO₄ (x=0, 0.3, 0.5, 0.7, and 1) powders were prepared via a high temperature solid state method [10]. The crystal structures of the obtained products were identified by XRD (X-ray diffraction) measurements using a powder X-ray diffractometer (PANalytical X′Pert³). The fluorescence spectra and decay curves were acquired using a fluorescence spectrometer (HORIBA Florolog-3) equipped with a 450 W xenon lamp and a pulsed 370 nm spectraLED (HORIBA). The temperature control of the sample was realized through heat conduction with a copper plate whose temperature was controlled using a heating tube and a temperature controller (OMRON E5CC-800).

3. Results and discussion

The XRD patterns of YV_1-xP_xO₄:3% Tb³⁺, 0.2% Eu³⁺ (x = 0, 0.3, 0.5, 0.7, and 1) powders are given in Fig. 1. The diffraction peaks of the as-prepared samples with x = 0 and 1 are in accord with the standard data of tetragonal YVO₄ (JCPDS No. 17-341) and YPO₄ (JCPDS No. 11-254), respectively, indicating that pure crystalline phase was obtained. What’s more, with the increase of P concentration from x = 0 to 1, the positions of diffraction peaks shift slightly toward larger angle direction because of the substitution of V⁵⁺ ions (ionic radius 0.59 Å) by smaller P⁵⁺ ions (ionic radius 0.34 Å) [16]. No additional diffraction peak exits in all of the samples. These indicate that a well-formed solid solution of YV_1-xP_xO₄ structure was obtained.

 

Fig. 1. XRD patterns of the as-prepared YV_1-xP_xO₄:3% Tb³⁺, 0.2% Eu³⁺ powder samples and the standard data of YVO₄ (JCPDS No. 17-341) and YPO₄ (JCPDS No. 11-254).

Download Full Size | PDF

In order to study the fluorescent properties of Tb³⁺ and Eu³⁺ in the solid solutions, emission spectra of the as-prepared samples were recorded under the excitation of 352 nm as shown in Fig. 2. This excitation wavelength was selected in order to guarantee that both Tb³⁺ and Eu³⁺ can be excited simultaneously. Specifically, under 352 nm excitation, Tb³⁺ can be excited from ground state ⁷F₆ to ⁵D₂, ⁵L₉, ⁵G₄ states [12]. In the case of Eu³⁺, because 352 nm locates in the tail of the V-O CTB, it can be excited due to the efficient energy transfer from the CT state.

 

Fig. 2. Normalized emission spectra of the as-prepared samples under the excitation of 352 nm (normalized to Eu³⁺ 620 nm emission).

Download Full Size | PDF

From the normalized emission spectra in Fig. 2, we found that the introduction of P into YVO₄ lattice is in favor of Tb³⁺ emission as expected, especially when the P concentration is larger than fifty percent. The detailed transitions corresponding to the emission peaks have been marked in Fig. 2 [12]. Moreover, it is observed that the spectral profiles of Eu³⁺ around 593 and 620 nm also change with the increase of P concentration. This is another evidence for the formation of solid solution YV_1-xP_xO₄. These variations of Tb³⁺ and Eu³⁺ emissions are ascribed to the variations of microstructures surrounding rare earth ions caused by the change in composition, which are irrelevant to the morphology of the as-prepared materials [17].

Excitation spectra of YVO₄:3% Tb³⁺, 0.2% Eu³⁺ and YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ monitored at Tb³⁺ 544 nm emission and Eu³⁺ 620 nm emission are shown in Figs. 3(a) and 3(b), respectively. The broad excitation band is ascribed to the V-O CT process. The sharp lines are ascribed to the characteristic f-f transitions of Tb³⁺ and Eu³⁺. For Eu³⁺ emission, the CT process is always efficient whether P⁵⁺ ions are introduced. But, in the case of Tb³⁺ emission in YVO₄:3% Tb³⁺, 0.2% Eu³⁺, no efficient excitation exists for both the CT process and the characteristic f-f transitions, which is due to the strong quenching of Tb³⁺ emission by Tb-V interaction [13]. However, when V⁵⁺ ions are partly substituted by P⁵⁺ ions, the separation between Tb³⁺ and V⁵⁺ increases, leading to a weakened Tb-V interaction. As a result, the quenching decreases and thus Tb³⁺ emission can be observed as shown in Fig. 2. When sufficient P were introduced, both the CT process and the characteristic f-f transitions turn efficient, which can be seen from Fig. 3(a).

 

Fig. 3. Excitation spectra of YV_1-xP_xO₄:3% Tb³⁺, 0.2% Eu³⁺ (x = 0 and 0.7) monitored at (a) Tb³⁺ 544 nm emission and (b) Eu³⁺ 620 nm emission.

Download Full Size | PDF

YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ sample was selected to study the temperature dependent fluorescence properties of Tb³⁺ and Eu³⁺ due to the strong emissions for both these two ions under 352 nm excitation. As shown in Fig. 4(a), Eu³⁺ emission increases dramatically when the temperature increases from 300 to 440 K. However, Tb³⁺ emission exhibits a contrary tendency. The insert in Fig. 4(a) shows such opposite variations quantificationally. I_541-555 and I_605-635 represent the integrated intensities for ⁵D₄ → ⁷F₅ transition (541-555 nm) of Tb³⁺ and ⁵D₀ → ⁷F₂ transition (605-635 nm) of Eu³⁺, respectively. The relationship between the ratio of I_605-635 to I_541-555 and temperature were phenomenologically fitted as shown in Fig. 4(b) using the following equation:

 

Fig. 4. (a) Emission spectra of YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ under excitation of 352 nm measured at different temperatures. (b) Temperature dependence of the FIR (I_605-635/I_541-555) of Eu³⁺ to Tb³⁺. (c) Relative sensitivity curve calculated from the fitting result in (b).

Download Full Size | PDF

The relative sensitivity S_R was obtained according to the fitting result which is shown in Fig. 4(c). The maximal S_R value reaches up to 2.7% K⁻¹ around 365 K.

To further confirm the feasibility of the as-proposed strategy, it was also performed using YV_0.3P_0.7O₄:3% Tb³⁺, 0.05% Eu³⁺ and YV_0.3P_0.7O₄:3% Tb³⁺, 0.1% Sm³⁺ samples. The temperature dependent emission spectra recorded under 352 nm excitation are shown in Fig. 5. Due to the decrease of Eu³⁺ concentration, the FIR of Tb³⁺ to Eu³⁺ is remarkably enhanced at a certain temperature, which can be seen from the comparison between Fig. 4(a) and Fig. 5(a).

 

Fig. 5. Emission spectra of (a) YV_0.3P_0.7O₄:3% Tb³⁺, 0.05% Eu³⁺ and (b) YV_0.3P_0.7O₄:3% Tb³⁺, 0.1% Sm³⁺ under excitation of 352 nm measured at different temperatures from 300 to 440 K.

Download Full Size | PDF

In the case of YV_0.3P_0.7O₄:3% Tb³⁺, 0.1% Sm³⁺ sample, four characteristic emission bands of Sm³⁺ centered at 564, 602, 646, and 704 nm can be observed, which are ascribed to ⁴G_5/2 → ⁶H_5/2, ⁴G_5/2 → ⁶H_7/2, ⁴G_5/2 → ⁶H_9/2, and ⁴G_5/2 → ⁶H_11/2 transitions, respectively [18]. The emission of Sm³⁺ exhibits similar temperature dependence with that of Eu³⁺ in the above studied materials, enhancing with the increase of temperature monotonously by virtue of the temperature-induced red shift of the CTB edge.

The integrated intensities I_541-555, I_605-635, and I_560-572 are used to describe the temperature dependent variations of Tb³⁺, Eu³⁺, and Sm³⁺ emissions, respectively, as shown in the insert of Fig. 5. The temperature dependences of the ratio of I_605-635 to I_541-555 in Fig. 5(a) and that of I_560-572 to I_541-555 in Fig. 5(b) were phenomenologically fitted using the following Eqs. (2) and (3):

The S_R values were calculated according to the formula in Fig. 4(c). The maximal S_R values for YV_0.3P_0.7O₄:3% Tb³⁺, 0.05% Eu³⁺ and YV_0.3P_0.7O₄:3% Tb³⁺, 0.1% Sm³⁺ samples reach up to 2.6% K⁻¹ around 380 K and 2.85% K⁻¹ around 365 K, respectively.

The mechanisms for temperature dependent variations of Eu³⁺/Sm³⁺ and Tb³⁺ were discussed as follows based on YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ sample taken as an example. Temperature dependent excitation spectra of YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ sample monitored at Eu³⁺ 697.5 nm and Tb³⁺ 544 nm emissions were recorded from 300 to 440 K as shown in Fig. 6 and Fig. 7, respectively. In the case of Eu³⁺, the CTB edge shifts to longer wavelength side with increasing temperature as expected, resulting in a dramatically enhanced excitation intensity at 352 nm, which can be seen from the enlarged image in Fig. 6. As for Tb³⁺, it was found that the excitation intensity decreases with increasing temperature. This is consistent with the temperature-induced decrease of Tb³⁺ emission mentioned above. In order to study whether the decrease of Tb³⁺ emission is related to the co-doped ions, emission spectra of Tb³⁺ single-doped YV_0.3P_0.7O₄ and Tb³⁺, Eu³⁺/Sm³⁺ co-doped YV_0.3P_0.7O₄ samples were measured under 352 nm excitation. The variations of the integrated emission intensity I_541-555 of Tb³⁺ with temperature are shown in Fig. 8 (the integrated intensities are normalized at 300 K). Obviously, the decreases of Tb³⁺ emission are similar for all the studied materials, indicating that the decrease is not related to the co-doped ions.

 

Fig. 6. Excitation spectra of YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ monitored at Eu³⁺ 697.5 nm emission measured at different temperatures.

Download Full Size | PDF

 

Fig. 7. Excitation spectra of YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺ monitored at Tb³⁺ 544 nm emission measured at different temperatures.

Download Full Size | PDF

 

Fig. 8. Normalized temperature-induced variations of Tb³⁺ emission intensity I_541-555 in Tb³⁺ single-doped and Tb³⁺, Eu³⁺/Sm³⁺ co-doped YV_0.3P_0.7O₄.

Download Full Size | PDF

Figure 9 presents the temperature dependent luminescence decay curves of Tb³⁺ 544 nm, Eu³⁺ 620 nm and Sm³⁺ 602 nm emissions of the three samples. It was found that the decay curves do not change with increasing temperature for all the cases, indicating that no temperature quenching occurs in the whole studied temperature range. Therefore, the temperature-induced decrease of Tb³⁺ emission can not be ascribed to temperature quenching behavior of Tb³⁺. We suggest that the increase of temperature may weaken the absorption of Tb³⁺, which further causes the decrease of Tb³⁺ emission.

 

Fig. 9. Temperature dependent decay curves of Tb³⁺ 544 nm, Eu³⁺ 620 nm and Sm³⁺ 602 nm emissions upon excitation of 370 nm in (a,b) YV_0.3P_0.7O₄:3% Tb³⁺, 0.2% Eu³⁺, (c,d) YV_0.3P_0.7O₄:3% Tb³⁺, 0.05% Eu³⁺, and (e,f) YV_0.3P_0.7O₄:3% Tb³⁺, 0.1% Sm³⁺.

Download Full Size | PDF

4. Conclusion

Based on the temperature-induced shift of the V-O CTB edge and temperature-induced variation of Tb³⁺ emission, ratiometric optical thermometry was designed in Tb³⁺, Eu³⁺/Sm³⁺ co-doped YV_1-xP_xO₄ solid solutions. The partial substitution of V by P played a crucial role for achieving and enhancing Tb³⁺ emission. This is caused by the weakened V-Tb interaction resulted from the increased separation between V⁵⁺ and Tb³⁺. Opposite temperature dependences for Eu³⁺/Sm³⁺ and Tb³⁺ emissions were achieved in YV_0.3P_0.7O₄:Tb³⁺, Eu³⁺/Sm³⁺ under 352 nm excitation in the range of 300-440 K. Specifically, Eu³⁺/Sm³⁺ emission increases dramatically with increasing temperature benefiting from the temperature-induced shift of the V-O CTB edge. While Tb³⁺ emission decreases monotonously. Using FIR of Eu³⁺/Sm³⁺ to Tb³⁺ as detecting parameter, high relative sensitivity was obtained, which is higher than 2% K⁻¹ in the range of 340-440 K. We conclude that this work may provide some new guidance in the development of optical temperature sensing as well as the application of rare earth doped luminescent materials.