Spectroscopic and Thermographic Qualities of Praseodymium-Doped Oxyfluorotellurite Glasses

The thermal stability of oxyfluorotellurite glass systems, (65-x)TeO2-20ZnF2-12PbO-3Nb2O5-xPr2O3, doped with praseodymium was examined. The different concentrations of praseodymium oxide (x = 0.5 and 2 mol%) were applied to verify the thermal, optical and luminescence properties of the materials under study. The relatively high values of the Dietzel (ΔT) and Saad–Poulain (S or H′) thermal stability factors determined using a differential thermal analysis (DTA) indicate the good thermal stability of the glass matrix, which gradually improves with the content of the active dopant. The temperature dependence of optical spectra in the temperature range 300–675 K for the VIS–NIR region was investigated. The involved Pr3+ optical transition intensities and relaxation dynamic of the praseodymium luminescent level were determined. The ultrashort femtosecond pulses were utilized to examine a dynamic relaxation of the praseodymium luminescent levels. Although the measured emission of the Pr3+ active ions in the studied glass encompasses the quite broad spectral region, the observed luminescence may only be attributed to 3PJ excited states. As a result, the observed decrease in the experimental lifetime for the 3P0 level along with the increasing activator content was identified as an intensification of the Pr–Pr interplay and the associated self-quenching process. The maximum relative sensitivities (Sr) estimated over a relatively wide temperature range are ~0.46% K−1 (at 300 K) for FIR (I530/I497) and 0.20% K−1 (at 600 K) for FIR (I630/I497), which seems to confirm the possibility of using investigated glasses in optical temperature sensors.

Moreover, in many cases, due to their unique properties as a host material, RE-doped glasses seem to be even more suitable than crystals.Glass production is easy and cheap compared to crystal.In addition, the glasses are more transparent, characterized by a high refractive index, good RE solubility, thermal stability, low phonon energy, good mechanical strength and proper non-optical linearity.They allow a relatively free selection of the chemical composition and/or its modification by various types of admixtures.This gives the possibility to influence the optical, fluorescence and melting point properties and the immediate surroundings of RE ions to a certain grade.From this point of view, it is known that the luminescence energy efficiency of rare earth ions increases if the glass matrix has low phonon energy, which is the case with tellurium glasses ≤ 800 cm −1 or fluoride glasses at ~500 cm −1 [6,19].The main purpose is therefore to create a glass matrix whose structure will provide an appropriate environment for rare earth element admixtures to obtain the best optical output of materials depending on their applications.In this context, fluorotellurite glass combines the advantages of fluoride glass (low-energy phonon environments) with the advantages of tellurite glass (chemical durability, thermal stability and mechanical strength) [6,[20][21][22].
Amongst the rare earth elements, praseodymium is a relatively alluring optical activator due to the amount of available energy levels in the UV-VIS-NIR range.Metastable states 1 D 2 and 3 P 0 , when stimulated, can emit red, green and blue light simultaneously for laser operation in both crystals [10,[23][24][25] and glass hosts [2,[26][27][28].And due to the 1 D 2 → 1 G 4 transition, Pr 3+ -doped glasses can also exhibit another interesting near-infrared (NIR) emission [29].Although many spectroscopic studies for various matrices/hosts (crystals, glasses, glass ceramics) activated with Pr 3+ ions have already been published, and optical enhancement based on 1 G 4 → 3 H 5 and 3 F 3,4 → 3 H 4 transitions in the NIR has been demonstrated, a description of the optical properties of Pr 3+ single-doped fluorotellurite glass is still relatively rare.
Therefore, this work characterizes the variability in the spectroscopic properties of oxyfluorotellurite glass (65-x)TeO 2 -20ZnF 2 -12PbO-3Nb 2 O 5 -xPr 2 O 3 under changes in the concentration (x = 0.5 and 2 mol%) of Pr 3+ ions and temperature (300-675 K).The designation TZPN was adopted for the base undoped glass 65TeO 2 -20ZnF 2 -12PbO-3Nb 2 O 5 .For glasses doped with praseodymium, the designations TZPN:0.5%Pr and TZPN:2%Pr were adopted, respectively.To the best of our knowledge, the glass composition that we propose, namely (65-x)TeO 2 -20ZnF 2 -12PbO-3Nb 2 O 5 -xPr 2 O 3 doped with praseodymium, has not yet been synthesized and/or examined.This type of amorphous material may be highly relevant owing to its still moderate red component of luminophores utilized in the light sources.The intentions of the present study are to (a) examine the impact of Pr 2 O 3 on the oxyfluoride glass thermal properties, (b) estimate the radiative transition rates based on the modified Judd-Ofelt phenomenological theory, and (c) determine the effect of temperature on absorption and emission spectra, eventually determining the related glass thermographic qualities.Moreover, the interionic peculiarities and dynamic relaxation of the luminescent excited state were studied by employing femtosecond laser pulses.

Thermal Features
In the initial phase of the research, the thermogravimetric measurements were carried out.DTA curves were recorded for the TZPN:0.5%Prand TZPN:2%Pr samples (Figure 1) and using the method of Keavney and Eberlin [30], characteristic temperatures were determined, such as the glass transition temperature (T g ) and onset of crystallization temperatures (T c ), in order to determine the thermal stability of the glasses.From the estimated data, it can be seen that the value of T g indicating the initiation of the glass softening increases slightly with the concentration of Pr 2 O 3 from 370.6 • C to 378.0 • C for samples containing 0.5 and 2 mol% Pr 2 O 3 , respectively.The T g value for TZPN undoped glass is even lower and equals 365.3 • C [31].In the case of the glass crystallization temperature, we also observe an increase in T c with the Pr 2 O 3 concentration rising.As it is seen in Figure 1, there is a shift from 529.5 • C with a 0.5 mole ratio of Pr 2 O 3 to 559.3 • C with a 2 mole ratio of Pr 2 O 3 .
It should be noted that the exothermic crystallization peak (T pc ) of both samples occurs in the range of 600-700 • C. Furthermore, in this case, we observe an increase in the T pc with the content of the active component.A similar effect has been observed in other glass systems [9,21,[32][33][34].In the case of undoped glass 65TeO 2 -20ZnF 2 -12PbO-3Nb 2 O 5 , values of the characteristic crystallization temperatures are T c = 552.3• C and T pc = 588.5 • C, respectively [31].All this affects the ability to form the glass and its thermal stability, which can be qualitatively determined by the criteria of thermal stability: Higher values of the parameters ∆T [30], H ′ [35], and S [35] mean better thermal stability of the glass and in the case of the investigated samples, their values are, respectively, TZPN:0.5%Pr(∆T = 158.9As it can be seen, thermal stability of the TZPN:Pr glasses improves with the content of Pr 2 O 3 .It should be noted that the estimation of the maximum T pc value may be subject to a certain (difficult to eliminate) error due to the asymmetric shape of the crystallization bands (a large number of physicochemical processes take place in this temperature range), dependence location of the crystallization peak of the technical parameters of the experiment (sample mass, heating rate) and constructional features of the apparatus [36].However, as shown in the obtained results, it does not have a major impact on the nature of changes in the thermal stability parameter S of the tested glass (it is the same trend as in the case of the parameters ∆T and H ′ ).It should be noted that the exothermic crystallization peak (Tpc) of both samples occurs in the range of 600-700 °C.Furthermore, in this case, we observe an increase in the Tpc with the content of the active component.A similar effect has been observed in other glass systems [9,21,[32][33][34].In the case of undoped glass 65TeO2-20ZnF2-12PbO-3Nb2O5, values of the characteristic crystallization temperatures are Tc = 552.3°C and Tpc = 588.5 °C, respectively [31].All this affects the ability to form the glass and its thermal stability, which can be qualitatively determined by the criteria of thermal stability:

Absorption Spectra and Modified Judd-Ofelt Analysis
Absorption spectra of TZPN:Pr glass samples were investigated in the wide UV-VIS-NIR range.Figure 2 shows the absorption spectrum taken at room temperature for the 63TeO 2 -20ZnF 2 -12PbO-3Nb 2 O 5 -xPr 2 O 3 glass sample in the wavenumber spectral range from 5000 to 25,000 cm −1 .The visible absorption bands on the left part of Figure 2 are related to the electron transitions of praseodymium from the ground state 3 H 4 to respective excited states: 3 H 6 and 3 F 2 (5133 cm −1 ); 3 F 3,4 (6595 cm −1 ); 1 G 4 (9851 cm −1 ); 1 D 2 (16,882 cm −1 ) and groups of bands 3 P 0,1,2 and 1 I 6 (21,773 cm −1 ).NIR range.Figure 2 shows the absorption spectrum taken at room temperature for the 63TeO2-20ZnF2-12PbO-3Nb2O5-xPr2O3 glass sample in the wavenumber spectral range from 5000 to 25,000 cm −1 .The visible absorption bands on the left part of Figure 2 are related to the electron transitions of praseodymium from the ground state 3 H4 to respective excited states: 3 H6 and 3 F2 (5133 cm −1 ); 3 F3,4 (6595 cm −1 ); 1 G4 (9851 cm −1 ); 1 D2 (16,882 cm −1 ) and groups of bands 3 P0,1,2 and 1 I6 (21,773 cm −1 ).The presented absorption spectrum was used to calculate the intensity of transitions based on the Judd-Ofelt theory [37,38].The application of the J-O theory in the case of the Pr 3+ ion as an optical activator is not so simple and depends on the type of host matrix [10].The absorption transitions satisfying the selection rules ΔS = 0, ΔL ≤ ±2 and ΔJ ≤ ±2 are the so-called hypersensitive transitions characterized by high values of experimental oscillator strengths [28].The values of experimental oscillator strengths presented in Table 1 indicate that the 3 H4 → 3 P2 transition is hypersensitive.
Moreover, the small energy difference between the ground state configuration 4f 2 and the first opposite parity excited configuration 4f 1 5d 1 results in a large deviation between the measured and calculated oscillator strengths and causes some problems for fitting 3 H4 → 3 P2 hypersensitive transition [6].To overcome these problems, a modified J-O theory should be used to estimate the phenomenological parameters Ωt.The presented absorption spectrum was used to calculate the intensity of transitions based on the Judd-Ofelt theory [37,38].The application of the J-O theory in the case of the Pr 3+ ion as an optical activator is not so simple and depends on the type of host matrix [10].The absorption transitions satisfying the selection rules ∆S = 0, ∆L ≤ ±2 and ∆J ≤ ±2 are the so-called hypersensitive transitions characterized by high values of experimental oscillator strengths [28].The values of experimental oscillator strengths presented in Table 1 indicate that the 3 H 4 → 3 P 2 transition is hypersensitive.Moreover, the small energy difference between the ground state configuration 4f 2 and the first opposite parity excited configuration 4f 1 5d 1 results in a large deviation between the measured and calculated oscillator strengths and causes some problems for fitting 3 H 4 → 3 P 2 hypersensitive transition [6].To overcome these problems, a modified J-O theory should be used to estimate the phenomenological parameters Ω t .
Following the standard J-O procedure, the theoretical oscillator strengths can be calculated using the following formula: where m is the electron mass, c is the speed of light, h is Planck's constant, λ is the mean wavelength of transition, (2J + 1) is the degeneracy of the ground state of the lanthanide ion, f N [L, S]J∥U t ∥ f N [L ′ , S ′ ]J ′ are double reduced matrix elements of the unit tensor and n is the refractive index.
On the other hand, the experimental oscillator strengths were estimated from the absorption spectrum using the relationship P exp = 4.318 × 10 −9 ε(ν)dν (5) where ε(ν) denotes the molar extinction and ν denotes the energy expressed in wavenumbers.Of course, the contribution of magnetic dipole transitions should also be taken into account and be subtracted from P exp before the Judd-Ofelt treatment.In the case of Pr 3+ ions, these are 3 H 4 → 3 F 3,4 and 3 H 4 → 1 G 4 transitions, respectively.Ultimately, using the least squares fitting method, this leads to the estimation of the three phenomenological parameters Ω 2,4,6 and after obtaining them, it leads to calculating the transition rates between any given states: where e is the charge of the electron and all other variables have the same meaning as in the previous equations.Next, one can obtain the values of luminescence branching coefficients and radiative lifetimes using the appropriate equations: Unfortunately, the proposed approach in the case of praseodymium quite often, and regardless of the type of host matrix, leads to a negative value of Ω 2 [6,10,28,[39][40][41].This requires a nonstandard approach and a certain modification of the J-O theory proposed by Kornienko [42] involving the following formula to calculate the intensity of electric dipoles: where α = ½ × [E4f5d − Ef 0 ] is called the fitting parameter, which in the case of Pr 3+ ions in any glass matrix is of the order of 1.0 × 10 −5 cm −1 ; P ′ cal is the oscillator force calculated from Equation ( 9); E J is the energy of the initial state; E J ′ is the energy of the final state; and Ef 0 is the energy corresponding to the center of gravity of the configuration, which is ~9940 cm −1 .
In the modified J-O model, the host-insensitive double reduced matrix elements (∥ U t ∥ 2 ) from standard J-O theory were multiplied by the fitting parameter α.The theoretical oscillator strengths calculated in this way (P ′ cal ) and experimentally determined (P exp ) are listed in Table 1.The estimated phenomenological J-O intensity parameters (Ω 2,4,6 ) are following: The relatively small root mean square deviation (RMS = 1.34 × 10 −6 ) indicates a good fit of P ′ cal with P exp and the optimal set of obtained intensity parameters: Ω 2 = 9.30 × 10 −20 cm 2 , Ω 4 = 13.24 × 10 −20 cm 2 and Ω 6 = 10.01 × 10 −20 cm 2 .As it is known, there is a certain relationship between the magnitude of the phenomenological parameters Ω 2,4,6 and the covalence of chemical bonds, structural changes in the vicinity of the incorporated rare earth ions (RE) and the bulk properties of the glass matrix [43].There is a higher value of the Ω 2 parameter than the greater degree of asymmetry around the Pr 3+ ions and stronger covalence of the Pr-O bonds [6].In turn, higher values of Ω 4 and Ω 6 mean lower rigidity and viscosity of the host material [10].The trend of J-O intensity parameters observed in the TZPN:Pr glass system is Ω 2 < Ω 6 < Ω 4 and is the same as in the case of other tellurite glasses doped with praseodymium [44][45][46][47].These values are significantly higher, indicating a more asymmetric and covalent environment around Pr 3+ ions in the tested system.In turn, the value of Ω 6 depends more on the overlap integrals of the 4f and 5d orbits than on the environment in which the Pr 3+ ions are located [6].From the absorption spectra (Figure 2-right) one can notice that as the temperature increases, the absorbance of the observed absorption transitions slightly decreases without changing their position.This indicates the homogeneous distribution of Pr 3+ ions in the glass TZPN:2%Pr and consequently, the excitation efficiency of the material under study may be insignificantly affected by the temperature elevation.

Emission Spectra, Radiative Properties and Color Perception
Using J-O parameters Ω 2,4,6 , the radiative properties of fluorescent transitions from 1 G 4 , 1 D 2 and 3 P 0 levels of the studied glasses are determined.The emission performance such as radiative transition probabilities (Wr), luminescence branching ratios (β) and radiative lifetimes (τ rad ) of the mentioned excited levels for the TZPN:2%Pr glass was estimated and collected in Table 2.The results of the J-O theory analysis fully confirm the emission spectra recorded at room temperature for TZPN:0.5%Prand TZPN:2%Pr glass samples.Typically, with 450 nm excitation, the Pr 3+ ions from the 3 H 4 ground level are excited to a higher-energy 3 P 2 excited state but as a result of non-radiative relaxation, a transition to the lower excited energy level 3 P 0 occurs.The excitation wavelength 450 nm related to a prominent 3 H 4 → 3 P 2 absorption line was selected to avoid a potential effect of the competitive processes, which might be harmful for the desired praseodymium luminescence.
Due to a small energy difference between the adjacent luminescent states 3 P 0 and 3 P 1 , they undergo thermalization and radiative relaxation has its source in both of these excited states ( 3 P 0,1 ).Therefore, the emission spectra presented in Figure 3 contain bands originating mainly from the 3 P 0 and 3 P 1 excited states.In the wavelength range from 450 nm to 1100 nm, 6-7 prominent bands can be seen, where the most intense are two bands corresponding to the transitions 3 P 0 → 3 H 4 (489 nm) and 3 P 0 → 3 F 2 (645 nm).
energy level 3 P0 occurs.The excitation wavelength 450 nm related to a prominent 3 H4 → 3 P2 absorption line was selected to avoid a potential effect of the competitive processes, which might be harmful for the desired praseodymium luminescence.
Due to a small energy difference between the adjacent luminescent states 3 P0 and 3 P1, they undergo thermalization and radiative relaxation has its source in both of these excited states ( 3 P0,1).Therefore, the emission spectra presented in Figure 3 contain bands originating mainly from the 3 P0 and 3 P1 excited states.In the wavelength range from 450 nm to 1100 nm, 6-7 prominent bands can be seen, where the most intense are two bands corresponding to the transitions 3 P0 → 3 H4 (489 nm) and 3 P0 → 3 F2 (645 nm).The influence of temperature on the emission spectrum of TZPN:0.5%Prglass in the visible and near-infrared range was also examined (see Figure 4).The emission intensity of all bands decreases with increasing temperature but the spectral shape and peaks' position are unaffected.The exception is that the band at 1327 nm corresponding to the 1 G4 → 3 H5 transition for that emission intensity diminishes with temperature increasing from 300 K to 600 K.The NIR luminescence at ~1.3 µm is widely used in telecommunications technology in the case of low-loss optical amplifiers operating in the O-band (1260-1360 nm) [48].The attenuation of conventional O-band optical fiber is relatively large, for example, almost twice that of the C-band (1530-1565 nm) [49].To compensate for this attenuation and effectively amplify light in the O-band, a praseodymium-doped fiber amplifier (PDFA) can be used based on a low-phonon-energy host glass and as one can see, the studied TZPN oxyfluorotellurite glass meets these requirements.The influence of temperature on the emission spectrum of TZPN:0.5%Prglass in the visible and near-infrared range was also examined (see Figure 4).The emission intensity of all bands decreases with increasing temperature but the spectral shape and peaks' position are unaffected.The exception is that the band at 1327 nm corresponding to the 1 G 4 → 3 H 5 transition for that emission intensity diminishes with temperature increasing from 300 K to 600 K.The NIR luminescence at ~1.3 µm is widely used in telecommunications technology in the case of low-loss optical amplifiers operating in the O-band (1260-1360 nm) [48].The attenuation of conventional O-band optical fiber is relatively large, for example, almost twice that of the C-band (1530-1565 nm) [49].To compensate for this attenuation and effectively amplify light in the O-band, a praseodymium-doped fiber amplifier (PDFA) can be used based on a low-phonon-energy host glass and as one can see, the studied TZPN oxyfluorotellurite glass meets these requirements.To determine the chromaticity of the emitted luminescence, the emission spectra profiles of TZPN glass doped with Pr 3+ were used.The chromaticity color coordinates estimated for the TZPN:0.5%Prglass sample excited at 450 nm are x = 0.490 and y = 0.385, respectively, and lie in the orange-red region of the chromaticity diagram of Commission Internationale de l'Eclairage (CIE) 1931 (see inset Figure 3).
The color temperature (CCT) correlated with them, determined based on the empirical McCamy formula [50]: To determine the chromaticity of the emitted luminescence, the emission spectra profiles of TZPN glass doped with Pr 3+ were used.The chromaticity color coordinates estimated for the TZPN:0.5%Prglass sample excited at 450 nm are x = 0.490 and y = 0.385, respectively, and lie in the orange-red region of the chromaticity diagram of Commission Internationale de l'Eclairage (CIE) 1931 (see inset Figure 3).
The color temperature (CCT) correlated with them, determined based on the empirical McCamy formula [50]: CCT = −449n 3 + 3525n 2 − 6823n + 5520.33 (10) where n = (x − x e )/(y − y e ) is the inverse slope line and x e = 0.3320 and y e = 0.1858 is the epicenter estimated based on the chromaticity coordinates.The CCT of TZPN:Pr oxyfluorotellurite glass at λ exc = 450 nm was found to be 2128.83K and therefore may be a promising candidate for the production of various types of photonic equipment (LEDs, solid-state lasers, displays and devices) emitting in the orange-red range of visible light [51,52].Another important parameter due to possible luminescent applications is color purity (CP), which determines how monochromatic/pure the emitted light is and was calculated according to the relationship where (x, y) are the estimated chromaticity color coordinates, (x s , y s ) are the standard coordinates of white light and (x d , y d ) are the coordinates of the dominant wavelength.The color purity for the glass sample TZPN:0.5%Prreached an effective value of 45.85%.Similar CP values are reported by Poojha et al. for Pr 3+ -doped lead boro-tellurite glasses [52].

Photoluminescence Decay Analysis
The images taken from the streak camera shown in the upper part of Figure 5 consist of luminescence originating in the 3 P 0 excited state.Examinations of photoluminescence (PL) decay curves for the 3 P0 excited lev lower part of Figure 5) show a partly single exponential character of the decay cur the TZPN:0.5%Prglass sample with a lifetime of 3 µs.When the concentration of Pr increases to 2 mol%, the decay becomes faster (τavg = 0.4 µs) and the decay curve de from a single exponential time dependence.Consequently, we follow a commonly a approximation and determine the so-called mean lifetime value expressed as follow  • () Examinations of photoluminescence (PL) decay curves for the 3 P 0 excited level (see lower part of Figure 5) show a partly single exponential character of the decay curve for the TZPN:0.5%Prglass sample with a lifetime of 3 µs.When the concentration of Pr 3+ ions increases to 2 mol%, the decay becomes faster (τ avg = 0.4 µs) and the decay curve deviates from a single exponential time dependence.Consequently, we follow a commonly applied approximation and determine the so-called mean lifetime value expressed as follows [53]: where t represents time and I is the intensity of luminescence.
The decrease in the experimental lifetime with a higher sample concentration is related to the self-quenching effect of Pr 3+ luminescence.This happens due to the reduced distance between Pr 3+ ions, which in turn results in the share of non-radiative energy transfer increases through cross-relaxation and energy migration processes.
At the same time, the obtained lifetimes of the excited state 3 P 0 are relatively shorter than a radiative lifetime obtained by the Judd-Ofelt calculation (see Table 2).The quantum efficiency (η) of the tested material estimated at over 62% results from the relationship and in many cases is higher or comparable to the corresponding data obtained for other praseodymium-doped glasses [6,10,27,28,52,54].
The Inokuti-Hirayama model can be applied to investigate the non-radiative Pr-Pr energy transfer in material under study [55].In the case of much slower energy migration in relation to interionic energy transfer, the time evolution of donor emission intensity may be defined as where Φ(t) is the emission intensity after pulse excitation, A is constant, S = 6 for dipole-dipole interactions, τ 0 is the intrinsic decay probability of the donor-involved excited state when the acceptor is absent and α is a parameter expressed as where Na denotes the acceptor concentration (Na = 2.36 × 10 21 ), R 0 is the critical Pr-Pr energy transfer distance and Γ = 1.77 (for S = 6) is Euler's function.In fact, the recorded 3 P 0 decay curve for TZPN:2%Pr glass is non-exponential and as a result of the appropriate fitting, we acquired the reasonable results for S = 6 and α = 5.64.Consequently, the critical energy transfer distance was estimated to be R 0 = 6.85 Å.An interaction parameter between praseodymium ions can be expressed as C da = R 0 6 •τ 0 −1 and the related value of 2.29 × 10 −38 cm 6 s −1 was estimated for our glass.Based on the formula W da = Cd a •R 0 −6 , the donor-acceptor energy transfer rate was found to be 2.49 × 10 6 s −1 .

Temperature Sensor Applications
In order to assess the suitability of the studied materials for applications in optical sensing thermometry, we examined changes in the fluorescence intensity ratio (FIR) as a function of temperature (in the range of 250-700 K) and corresponding absolute (S A ) and relative (S R ) thermal sensitivities for TZPN:0.5%Prglass.
The obtained results are presented in Figure 6 in the form of temperature dependence curves of fluorescence intensity ratios FIR (I 530 /I 497 ) (on the left) and FIR (I 630 /I 497 ) (on the right) by fitting with the following relationship: where ∆E is the energy difference between the thermalized levels, k B is the Boltzmann constant, T is the temperature expressed in the absolute scale [K] and A and B are constants.
In the case of FIR (I 530 /I 497 ), we observe a linear increase in the fluorescence intensity ratio with a maximum absolute temperature sensitivity value of S A = 5.1 × 10 −3 K −1 at T = 460 K and approximately 0.46% K −1 relative temperature sensitivity for T = 300 K.The values of the fluorescence intensity ratios FIR (I 630 /I 497 ) increase exponentially with temperature and the maximum absolute temperature sensitivity is reached for T = 675 K and amounted to 8.7 × 10 −3 K −1 , which translates into S R = 0.20% K −1 of relative temperature sensitivity at about T = 460 K.All absolute and relative curves of temperature sensitivity for TZPN:0.5%Prglass have a non-linear way with a different tendency.For a comparison, A.S. Rao examined the impact of temperature on four different emission bands of praseodymium [56].Some fluorescence intensity ratio (FIR) models based on the relationships between different emission peaks were studied to estimate a maximum relative sensitivity, 1.03% K −1 .Moreover, other Pr-doped inorganic phosphors were investigated by Jiawen Wang et al. [57] and a comparable evaluation approach gave rise to quite high relative sensitivity (1~3.25%K −1 ) and low temperature uncertainty (0.15-0.5 K).These reported sensitivities are higher in relation to TZPN:Pr estimations but regardless, our results are attributed to a wide temperature range up to 675 K.
The obtained results are presented in Figure 6 in the form of temperature dependence curves of fluorescence intensity ratios FIR (I530/I497) (on the left) and FIR (I630/I497) (on the right) by fitting with the following relationship: where ∆E is the energy difference between the thermalized levels, kB is the Boltzmann constant, T is the temperature expressed in the absolute scale [K] and A and B are constants.In the case of FIR (I530/I497), we observe a linear increase in the fluorescence intensity ratio with a maximum absolute temperature sensitivity value of SA = 5.1 × 10 −3 K −1 at T = 460 K and approximately 0.46% K −1 relative temperature sensitivity for T = 300 K.The values of the fluorescence intensity ratios FIR (I630/I497) increase exponentially with temperature and the maximum absolute temperature sensitivity is reached for T = 675 K and amounted to 8.7 × 10 −3 K −1 , which translates into SR = 0.20% K −1 of relative temperature sensitivity at about T = 460 K.All absolute and relative curves of temperature sensitivity for TZPN:0.5%Prglass have a non-linear way with a different tendency.For a comparison, A.S. Rao examined the impact of temperature on four different emission bands of praseodymium [56].Some fluorescence intensity ratio (FIR) models based on the relationships between different emission peaks were studied to estimate a maximum relative sensitivity, 1.03% K −1 .Moreover, other Pr-doped inorganic phosphors were investigated by Jiawen Wang et al. [57] and a comparable evaluation approach gave rise to quite high relative sensitivity (1~3.25%K −1 ) and low temperature uncertainty (0.15-0.5 K).These reported sensitivities are higher in relation to TZPN:Pr estimations but regardless, our results are attributed to a wide temperature range up to 675 K.

Materials and Methods
A traditional melt quenching method was employed to manufacture TeO 2 -ZnF 2 -PbO-Nb 2 O 5 oxyfluoride glasses activated with 0.5 and 2 mol% of Pr 2 O 3 .Tellurium oxide (5N), zinc fluoride (5N), lead oxide (4N), niobium oxide (Nb 2 O 3 ) and praseodymium oxide (5N) were thoroughly mixed, ground and incorporated in a corundum crucible.The melting process in the ambient atmosphere took 30 min at 830 • C. The glasses were poured into preheated brass molds and then annealed for a few hours at 350 • C in order to reduce the internal stresses.The resulting glass samples were transparent and homogeneous.
The specimens were characterized by the DTA (differential thermal analysis) technique applying the normal pressure and atmosphere.The DSC 404/3/F calorimeter (Erich NETZSCH B.V. & Co. Holding KG Gebrüder-Netzsch, Selb, Germany), platinum crucibles and reference holders were adequately employed.The same heating rates (10 K/min) were realized for all studied glasses.
The spectrophotometer Agilent Cary 5000 (Agilent, Santa Clara, CA, USA) was used to measure the survey absorption spectra within UV-VIS-NIR spectral ranges.Lumi-nescence experiments were carried out in the visible and near-infrared spectral regions as well.An FLS1000 Spectrofluorimeter (Edinburgh Instruments Ltd., Livingston, UK) was utilized to record the emission spectra.The Linkam THMS 600 Heating/Freezing Stage (Linkam Scientific Instruments Ltd., Redhill, UK) was used to perform temperaturedependent measurements.To record luminescence decay curves, the glass samples were excited by a femtosecond LIBRA Ti:sapphire laser (Coherent Inc., Santa Clara, CA, USA) coupled with optical parametric amplifier "Opera" (OPO).

Conclusions
It was found that TZPN glass thermal stability increases for a higher concentration of optically active ions.The impact of temperature on the absorption and variation of fluorescence intensity in the VIS-NIR region was monitored in the temperature range of 300-675 K.The observed luminescence can be mainly attributed to transitions originating from two closely located, thermally coupled levels ( 3 P 0 and 3 P 1 ).The values of the radiative transition probabilities Wr, branching ratios β and the fluorescence intensity ratio at different temperatures for the selected praseodymium transitions were adequately investigated.The branching ratio of praseodymium luminescence in material under study is related to a wide spectral region; hence, its various useful potential applications can be considered.The CIE coordinates for the TZPN:Pr glasses are in the orange-red region.The maximum lifetime of the 3 P 0 level was observed for a sample with a lower concentration of active ions.This is explained by the self-quenching effect-the experimental lifetime decreases with increasing Pr 3+ concentration.Application potential of the investigated material in optical sensor thermometry was evaluated as well.

Molecules 2024 ,
29, x FOR PEER REVIEW

Figure 6 .
Figure 6.Fluorescence intensity ratios attributed to praseodymium luminescence as well as corresponding absolute and relative thermal sensitivities estimated for FIR (I 530 /I 497 ) and FIR (I 630 /I 497 ) for TZPN:0.5%Prglass.