1. Introduction

In recent years, the optical temperature sensor behaviour based on the upconversion (UC) emission attracted much attention for their high sensitivity and accuracy [1]. Compared with the conventional temperature sensors that based on the principle of liquid and metal expansion, optical thermometry of UC luminescence based on the fluorescence intensity ratio (FIR) technique possess many unique advantages, such as contactless measurement, large-scale imaging and so on [2]. For this kind of UC phosphor materials, the temperature is usually measured by evaluating the change in luminescence observed from two thermally coupled levels of rare earth (RE) ion, such as Pr³⁺ (³P₀,³P₁→³H₅), Nd³⁺ (⁴F_3/2,³H₄→⁴I_9/2), Dy³⁺ (⁴F_9/2,⁴I_15/2→⁶H_13/2), Er³⁺ (⁴S_3/2,⁴F_9/2→⁴I_15/2) and so on [3]. In this work, besides the study of the conventional Yb³⁺-Er³⁺ couple for temperature sensing, it has been found that the FIR of the (⁵F₄,⁵S₂)-⁵I₈ to ⁵F₅-⁵I₈ transitions of Ho³⁺ is also changed with the temperature although the two levels are not thermally coupled levels. The corresponding mechanism was also discussed. The developed Yb³⁺-Ho³⁺ doped phosphor can exhibit excellent sensitivity for temperature sensing, which provides a new strategy for selecting RE ions to apply to temperature sensors.

So far, high quantum efficiency for UC luminescence is most obtained in fluoride phosphors. However, fluorides are very sensitive to oxygen surface contamination, which may influence the luminescence and limit the applications [4]. In contrast, oxides generally show high chemical stability and environmental-friendly characters. So, more and more oxide-based UC phosphors have been paid attention in recent years. It is known the apatite-type alkaline-earth halo phosphates with the general formula M₅(PO₄)₃Cl (M = Ca, Sr, Ba) are still unsurpassed as a family of phosphor matrix. For instance, the long lasting properties of the Sr₅(PO₄)₃F_xCl_1-x:Eu²⁺,Gd³⁺ were reported by Deng et al. [5]; Guo et al. studied the energy transfer (ET) between Eu²⁺ and Mn²⁺ in Sr₅(PO₄)₃Cl [6]. In this paper, to explore new luminescent materials for temperature sensing, the Yb³⁺-Ho³⁺ and Yb³⁺-Er³⁺ doped Sr₅(PO₄)₃Cl phosphors were designed, and their luminescence properties used for temperature sensing were investigated.

2. Experimental

The Sr_4.75(PO₄)₃Cl:0.1Yb³⁺,0.025Ho³⁺,0.125Na⁺ (abbreviated as SPC:Yb³⁺,Ho³⁺) and Sr_4.75(PO₄)₃Cl:0.1Yb³⁺,0.025Er³⁺,0.125Na⁺ (abbreviated as SPC:Yb³⁺,Er³⁺) samples were prepared by the solid-state reaction method. The starting materials included SrCO₃ (AR), Na₂CO₃ (AR), (NH₄)₂HPO₄ (AR), SrCl₂·6H₂O (AR), Yb₂O₃ (4N), Er₂O₃ (4N) and Ho₂O₃ (4N). Na₂CO₃ was used as the charge compensator. Stoichiometric amounts of the starting materials were mixed and ground together by an agate mortar, and then calcined at 1050 °C for 3.5 h. The phase purity was determined by an ARL X'TRA powder X-ray diffractometer (XRD) with Cu Kα radiation (λ = 1.5418 Å) operating at 40 kV and 35 mA. The UC spectra and decay curves were measured on an EI-FS5 fluorescence spectrophotometer system (Semiconductor refrigeration of R928P is used for the detector with the dark noise blew 100 cps) equipped with 980 nm laser diode (LD) and laser pulsed source (pulse width is 35 μs), respectively. For the temperature dependent measurement, the samples were mounted on a heating device whose temperature could be changed from room temperature to 573 K with the step of 0.1 K.

3. Results and discussion

Figure 1(a) presents the XRD patterns of SPC:Yb³⁺,Ho³⁺ and SPC:Yb³⁺,Er³⁺. All the diffraction peaks can be indexed to pure hexagonal-structured SPC (JCPDS Card No.83-0973). No obvious impurity phase was detected, revealing both the as-prepared samples are single-phase.

 

Fig. 1 XRD patterns of SPC:Yb³⁺,Er³⁺ and SPC:Yb³⁺,Ho³⁺.

Download Full Size | PDF

Figure 2(a) represents the emission spectra of SPC:Yb³⁺,Ho³⁺ under various LD excitation powers. Three emission peaks located around 543, 668 and 758 nm are found for every spectrum, which could be assigned to the (⁵F₄,⁵S₂)-⁵I₈, ⁵F₅-⁵I₈ and (⁵F₄,⁵S₂)-⁵I₇ transitions of Ho³⁺, respectively [7]. The Commission International de l’Eclairage (CIE) chromaticity coordinates were calculated to be (0.318, 0.674), revealing a yellowish green emission. With increasing excitation power, the spectral profile changes little but the emission intensity increases continuously. For unsaturated UC processes, the number of photons which are required to populate the upper emitting state can be obtained by the relation $I_f\propto P^n$, where I_f is the fluorescent intensity, P is the pump power, and n is the number of photons required to populate the emitting state [8]. The inset of Fig. 2(a) shows the dependence of the emission intensities on the excitation powder, where the n values were obtained to be 1.70, 1.61 and 1.38 for 543, 668 and 758 nm, respectively. Thus, the these emission are two-photon process.

 

Fig. 2 Emission spectra of (a) SPC:Yb³⁺,Ho³⁺ and (b) SPC:Yb³⁺,Er³⁺ under various LD excitation powers, inset shows the dependence of UC emission intensities on excitation powder.

Download Full Size | PDF

Figure 2(b) shows the emission spectra of SPC:Yb³⁺,Er³⁺ under various LD excitation powers. It can be seen that the spectral profile also changes little and the emission intensity increases with increasing excitation power. The main emission peaks are located in the region of 640-700 nm, attributed to the ⁴F_9/2-⁴I_15/2 transition of Er³⁺ [3]. It is worth noting that the green (²H_11/2,⁴S_3/2)-⁴I_15/2 transition peaks of Er³⁺ are very weak relative to the red ones, which could be observed in the enlarged emission spectra from 500 to 600 nm. As shown in Ref [9], the ⁴F_9/2 level of Er³⁺ in the UC process can be populated by nonradiative relaxation from the upper (²H_11/2, ⁴S_3/2) levels or the lower ⁴I_13/2 level which is initially populated via nonradiative ⁴I_11/2-⁴I_13/2 relaxation. In both the cases, the nonradiative relaxation play an important role. It is known that the oxide host shows a much higher phonon energy relative to the fluoride, which intensifies the above nonradiative relaxations and results in the strong red ⁴F_9/2-⁴I_15/2 emission in the SPC:Yb³⁺,Er³⁺. In the inset of Fig. 2(b), a plot of lnI_f versus P yields a straight line with slope n = 1.37 for the red emission from 640 to 700 nm, indicating a two-photon process.

Figure 3(a) presents the emission spectra of SPC:Yb³⁺,Ho³⁺ excited at 980 nm under various temperatures. It can be found the whole spectral intensity decreases with increasing temperature, but the green emission at 543 nm shows a faster decay in intensity relative to the red one at 668 nm. To further understand this point, Fig. 3(b) depicts their normalized (at 543 nm) emission spectra. With the temperature increased from 293 to 553 K, the intensity of the red emission increases obviously. As mentioned in the Introduction part, the green and red emissions of Ho³⁺ are not thermally coupled, but their intensity ratio depends on the temperature as witnessed in Fig. 3(b). The reason will be interpreted later. Figure 3(c) presents the dependence of FIR for the 668 and 543 nm emissions on the absolute temperature, where I₅₄₃ and I₆₆₈ are the integrated intensities of the Ho³⁺ (⁵F₄,⁵S₂)-⁵I₈ and ⁵F₅-⁵I₈ transitions, respectively. The FIR increases from 0.458 to 1.984 with increasing temperature from 293 to 553 K, respectively. The experimental data can be well fitted by a single-exponential formula $R=0.55\exp (T/339.7)-0.83$. For sensing application, the relative sensitivity S is a very important parameter which can be calculated as [3]

 

Fig. 3 (a) Emission spectra of SPC:Yb³⁺,Ho³⁺ at various temperatures; (b) normalized emission spectra of SPC:Yb³⁺,Ho³⁺ at various temperatures; (c) dependence of FIR for the 668 and 543 nm emissions of SPC:Yb³⁺,Ho³⁺ on the absolute temperature; (d) sensitivity as a function of the temperature of 293 to 553 K for SPC:Yb³⁺,Ho³⁺.

Download Full Size | PDF

Figure 4(a) shows the emission spectra of SPC:Yb³⁺,Er³⁺ excited at 980 nm at various temperatures. For the ⁴S_3/2-⁴I_15/2 transition of Er³⁺ around 544 nm, the emission intensity exhibits a continuous decrease with increasing temperature. But the Er^{3+ 2}H_11/2-⁴I_15/2 transition intensity demonstrates an increase till T = 453 K, and then shows a decrease. As a result, the FIR for the Er³⁺ 522 and 544 nm emissions will show a regular change with temperature. To interpret this, Fig. 4(b) presents the normalized (at 544 nm) emission spectra of SPC:Yb³⁺,Er³⁺ at various temperatures. As expected, the Er³⁺ emission intensity increases continuously with increasing temperature. This phenomenon is owing to the small energy gap (~750 cm⁻¹) between the two thermally coupled levels ²H_11/2 and ⁴S_3/2 of Er³⁺ [3]. The relative population of the two thermally coupled multiplets follows the Boltzmann distribution, written as [10,11]

 

Fig. 4 (a) Emission spectra of SPC:Yb³⁺,Er³⁺ at various temperatures; (b) normalized emission spectra of SPC:Yb³⁺,Er³⁺; (c) dependence of FIR for the 522 and 544 nm emissions on the absolute temperature; (d) sensitivity as a function of the temperature from 293 to 553 K.

Download Full Size | PDF

To interpret the different decay rates of the 543 and 668 nm emission intensities of Ho³⁺ by excitation under different temperatures in Fig. 3(a), Fig. 5(a) demonstrates the schematic representation of the energy level diagram for Yb³⁺-Ho³⁺ as well as the proposed UC mechanisms. The detailed UC ET processes have been discussed in Ref [9]. When the temperature increases, the lattice vibration of this compound is intensified, which increases the phonon number in the host. According to Ref [13], the non-radiative de-excitation probability (K_nr) can be approximately described by the Mott-Seitz model by $K_{nr}\propto \exp (-\frac{\Delta E}{{\kappa }_{B}T})$, where ΔE is the energy gap between two levels, κ_B is the Boltzmann constant and T is the absolute temperature. Thus, the increasing temperature could urge the occurrence of the nonradiative relaxations ① and ② in Fig. 5(a), which reduces the population number of the electrons on the ⁵F₄,⁵S₂ levels of Ho³⁺ and is beneficial to the electron population on the ⁵F₅ level. As a result, the intensity ratio of the red (668 nm) to green (543 nm) emissions has been enhanced with the temperature increased. Nevertheless, the red emission is still decreased somewhat when the temperature increases compared with that under room temperature condition due to the increasing nonradiative transition from the ⁵F₅ level at high temperature. To further verify the intensification of the nonradiative relaxation ②, Fig. 5(b) represents the decay curves of SPC:Yb³⁺,Ho³⁺ with 980 nm excitation and 543 nm emission under the temperatures of 293 and 443 K. Both the decay curves could be well-reproduced by a double-exponential function as $I=A_1\exp (-t/{\tau }_1)+A_2\exp (-t/{\tau }_2)$, where τ₁ and τ₂ are the fast and slow components of the luminescent lifetimes, A₁ and A₂ are the fitting parameters, respectively. The corresponding τ_i and A_i (i = 1, 2) values of the decay curves were summarized in Table 1. To determine the average lifetimes for each decay curve, the formula $<\tau >=(A_1{\tau }_1^2+A_2{\tau }_2^2)/(A_1{\tau }_1+A_2{\tau }_2)$ is employed [14]. The average decay lifetimes were calculated to be 18.08 and 4.79 μs for T = 293 and 443 K, respectively. The sharp decay of the Ho³⁺ lifetime with increasing temperature indicates that the energy at the ⁵F₄,⁵S₂ levels of Ho³⁺ has a faster transfer to the lower levels when the temperature is higher.

 

Fig. 5 (a) Schematic representation of the energy level diagram for Yb³⁺-Ho³⁺; (b) decay curves of SPC:Yb³⁺,Ho³⁺ under the temperatures of 293 and 433 K.

Download Full Size | PDF

Table 1. τ_i and A_i (i = 1, 2) values of the decay curves for SPC:Yb³⁺,Ho³⁺ with 980 nm excitation and 543 nm emission under the temperatures of 293 and 433 K

View Table

4. Conclusions

In this work, the SPC:Yb³⁺,Ho³⁺ and SPC:Yb³⁺,Er³⁺ phosphors were synthesized by the solid-state reaction method. The green and red emission peaks of Ho³⁺ are found around 543 and 668 nm, respectively. The Er³⁺ in the SPC host exhibits a very intense red emission around 673 nm compared with the green one from 515 to 575 nm. The UC mechanisms are elucidated through the laser power dependence of the upconverted emissions. The investigation of temperature-dependence reveals that the green and red emission intensities of Ho³⁺ have different decay rates, and thus a changing intensity ratio of the two emissions as a function of temperature could be obtained. The FIR of the 522 to 544 nm emissions of Er³⁺ also changes with the temperature owing to the two thermally coupled levels of ²H_11/2 and ⁴S_3/2 for Er³⁺. Their sensitivity of the as-prepared samples was also evaluated, showing high sensitivity.