Eu²⁺ emission from thermally coupled levels – new frontiers for ultrasensitive luminescence thermometry

Abstract

Designing luminescent thermometers operating in a broad T-range, spanning hundreds of degrees, presenting high thermal sensitivity and good resolution is one of the greatest challenges in the field. In this paper, we present SrB₄O₇:Eu²⁺, a novel luminescent thermometer, operating in the T-range of 11–600 K and presenting unprecedentedly high relative thermal sensitivity, i.e., up to ≈10.5%/K and ≈22.6%/K, using either Eu²⁺ luminescence decay time or intensity ratio as thermometric parameters, respectively. Such good performance classifies the SrB₄O₇:Eu²⁺ phosphor as the most sensitive inorganic luminescent thermometer ever reported. The temperature dependences of both luminescence intensity ratios and the emission decay time result from the same physical effect, i.e., thermal coupling of the lowest 5d₁ excited level of Eu²⁺ and the ⁶P_7/2 excited level of the 4f⁷ configuration of the dopant. The latter one is located ≈130 cm⁻¹ below the 5d₁. Such a strong coupling changes the contribution of the emission from the two levels, in response to the temperature change, but without any significant luminescence thermal quenching. This beneficial situation allows measuring spectra with a high signal-to-noise ratio, which lowers the uncertainty of the temperature measurement (improves resolution) in the whole operating T-range. The SrB₄O₇:Eu²⁺ material offers a broad operating T-range, as well as unprecedentedly high relative thermal sensitivity and low uncertainty of the temperature measurement.

---

1. Introduction

Developing highly sensitive luminescent thermometers is a challenging task. When the required/expected temperature range to be detected exceeds several hundred degrees, the design of such thermometers becomes even more demanding.^1,2 As noted in ref. 3 a wide measuring range and high sensitivity of luminescent thermometers are slightly opposite requirements. Hence, one cannot expect that the high relative thermal sensitivity may be retained over the range of hundreds of degrees. However, it is conceivable a luminescent thermometer can operate over such a wide T-range, offering simultaneously high thermal sensitivity in the whole T-range, due to the use of a multi-parameter (multi-mode) sensing strategy.^4–6 Such characteristics of a luminescent thermometer may be useful in various practical applications.

Luminescence of Eu²⁺ typically consists of broadband emission resulting from its 5d → 4f transition. Fig. 1a presents the situation when the 5d level has lower energy than the ⁶P_7/2 one and then only the 5d → 4f emission may occur. This situation is not attractive from the point of view of luminescence thermometry. Yet, Eu²⁺ may also generate photons as a consequence of the 4f → 4f transition, when the ⁶P_7/2 excited manifold of the 4f⁷ configuration is located below the lowest-energy 5d₁ level of the ion (see Fig. 1b). While generally uncommon, in some hosts such 4f → 4f emission of Eu²⁺ occurs, presenting the characteristic narrow lines, associated with the intra-configurational ⁶P_7/2 → ⁸S_7/2 transition.^7,8 For the reasons discussed below, it may occur at rather low temperature values (cryogenic T-range). If the ⁶P_7/2 level is located below the 5d₁ one, both levels may be thermally-coupled. Since the lifetime of the ⁶P_7/2 level is typically much longer (forbidden f–f transitions) than the lifetime of the 5d level (parity-allowed d–f transitions), once the coupling occurs, the 5d → 4f transition contributes to the emission effectively. The luminescence character may then change upon the increase of temperature, from the entirely intra-configurational 4f → 4f transition (narrow lines) to the inter-configurational 5d → 4f one (broadband).⁸

Fig. 1 Schematic diagram of the energy levels of the Eu²⁺ ion. In (a), the 5d level is positioned below the ⁶P_7/2 one, which precludes the ⁶P_7/2 → ⁸S_7/2 radiative relaxation, and only the 5d → 4f luminescence occurs. In (b), both narrow and broadbands corresponding to the ⁶P_7/2 → ⁸S_7/2 and 5d → 4f(⁸S_7/2) transitions, respectively, are possible, and their relative intensities (populations) are mainly governed by thermal coupling (situation valid for the SrB₄O₇:Eu²⁺ system).

In such a situation, with increasing temperature, not only does the luminescence spectrum change significantly but also the emission decay time shortens, as the contribution of the parity-allowed 5d → 4f transition increases.¹⁰ Consequently, both the shape of the luminescence spectrum and the decay time of the emission may be potentially useful for thermometric applications. We shall present in this paper that such characteristics of Eu²⁺ luminescence may lead to unprecedentedly high relative thermal sensitivity of an optical thermometer, combined with a very broad operating T-range. Importantly, we consider the harmonization of both effects within a single inorganic compound is a significant achievement with the prospect of searching for other phosphors showing similar effects, and possibly offering even better thermometric performance.

The SrB₄O₇ host lattice is an excellent candidate for activation with Ln²⁺ ions. It has been shown that the tetrahedral structure of the BO₄ groups stabilizes such dopants effectively, even in an oxidizing atmosphere and at high temperatures.^11,12 The reducing capability of the host has been already demonstrated for Sm²⁺,^13,14 Yb²⁺,¹⁵ Tm²⁺,¹⁶ Nd²⁺,¹⁷ and even Bi^2+18,19 ions, despite their lower reduction potential, compared to Eu³⁺. The stability of Eu²⁺ in the SrB₄O₇ host was proven many years ago.⁸

Recently, we have presented the benefits of applying the 5d → 4f emissions of Pr³⁺ and Tm²⁺ in luminescence thermometry, showing that the use of their inter-configurational transitions greatly enhances the relative thermal sensitivity of such thermometers.^20–24 Now, we turn our attention to the Eu²⁺ luminescence and its possible utility in the optical determination of temperature. The emission intensity of this ion is often strong, due to the broadband parity-allowed 4f → 5d excitation transition, that can be easily realized using standard light sources, which also widens the variety of potential applications.

Until now, a few strategies to exploit Eu²⁺ luminescence as an optical thermometer were employed. These involved the ratio of the Eu²⁺/Eu³⁺ emission intensities,^25,26 the ratio of emission bands from Eu²⁺ occupying different sites of the host lattice,^27,28 the rate of energy transfer between Eu²⁺ and other d- or f-block elements.²⁹ Yet, to the best of our knowledge, there are no reports on the use of the Eu²⁺ 5d → 4f and 4f → 4f transition intensity ratio as a thermometric parameter. Detailed analysis of the fundamental research results on SrB₄O₇:Eu²⁺⁸ led us to the conclusion that this phosphor may offer particularly interesting thermometric properties. In this paper, we explore the idea focusing both on the SrB₄O₇:Eu²⁺ luminescence spectra and the emission kinetics. We show that SrB₄O₇:Eu²⁺ may work as a dual-mode ratiometric luminescent thermometer, using the luminescence intensity ratio (LIR) of its emission features and decay time of its luminescence. Moreover, we prove that it can operate in a broad T-range and show unprecedentedly high relative thermal sensitivity in cryo-range, whichever of the two modes is exploited.

2. Experimental

2.1. Synthesis

A powder SrB₄O₇:Eu²⁺ (3 mol%) was synthesized via a conventional solid-state (ceramic) method. SrCO₃ (Sigma-Aldrich, 99.9%), H₃BO₃ (Chempur, p.a.), Eu₂O₃ (Stanford Materials, 99.99%) were the starting materials. The stoichiometric mixture of the reagents (with 5 mol% excess of H₃BO₃) was thoughtfully ground in an agate mortar with some acetone as a wetting medium. After natural drying, the mixture was transferred into a porcelain crucible covered with a lid and was heated in a furnace for 5 h at 700 °C in air. After cooling down, the powder was ground again and heat-treated for 5 h at 850 °C twice, with an intermediate grinding. It was crucial to cover the crucible with a lid to preclude the evaporation of B₂O₃ from the reacting mixture and limit the oxygen flow.

2.2. Characterization

Powder X-ray diffraction pattern (XRD) was recorded using a Bruker D8 Advance diffractometer. The measurement was performed in the 2θ range of 10–60°, using Cu Ka₁ radiation (λ = 1.5406 Å) with the counting time 0.2 s and the step 0.008°. The reference pattern was taken from Inorganic Crystal Structure Database (ICSD, #14383). The morphology of the sample was analyzed using a Hitachi S-3400N scanning electron microscope (SEM). Photoluminescence (PL), PL excitation spectra (PLE), and luminescence decay (DEC) traces were recorded in the 11–600 K temperature range, with FLS1000 Fluorescence Spectrometer from Edinburgh Instruments Ltd. For PL and PLE measurements a 450 W continuous xenon arc lamp was used as the excitation source, while for DEC traces a 60 W Xe pulse lamp was employed. The sample was mounted on a copper holder in a closed-cycle helium cryostat from Lake Shore Cryotronics, using Silver Adhesive 503 glue from Electron Microscopy Sciences. TMS302-X double grating excitation and emission monochromators of 325 mm focal lengths were used, and the luminescence signal was recorded with a Hamamatsu R928P high-gain photomultiplier detector, thermoelectrically cooled to −22 °C. The excitation spectra were corrected for the incident light intensity and the emission spectra were corrected for the spectral sensitivity of the recording system.

3. Results

3.1. Structural and morphological properties.

Fig. 2a–c presents the results of the structural and morphological characterization of the obtained powder phosphor. The measured XRD diffractogram matches perfectly with the ICSD #14383 reference pattern of SrB₄O₇, proving the expected orthorhombic structure (P2₁nm) of the synthesized phosphor material. In SrB₄O₇ there is only one Sr²⁺ cationic site with C_s point symmetry, and Eu²⁺ obviously enters this position. Representative SEM images of the synthesized phosphor material are presented in Fig. 2b and c. The grains have a narrow size distribution and show an elongated shape with about 1 μm length and about 100–150 nm thickness.

Fig. 2 (a) X-Ray diffraction pattern of SrB₄O₇ containing 3% of Eu²⁺ and prepared in an air atmosphere. (b and c) SEM images of the SrB₄O₇:Eu²⁺ powder, taken at different magnifications (b - 5 μm scale bar; c - 2 μm scale bar).

3.2. Photoluminescence properties

Fig. 3a presents the PLE spectra of Eu²⁺ luminescence monitored at 367.2 nm and recorded at 11 and 300 K. Fig. 3b shows the PL spectra recorded at the same temperature values, under 300 nm excitation. A detailed analysis of the luminescence spectroscopy of the SrB₄O₇:Eu²⁺ system, including assignment of the observed lines, has already been presented by Meijerink et al.⁸ Here, we only discuss the temperature-dependent properties, important for the thermometric analysis presented below. The PLE spectra (Fig. 3a) consist of two broad overlapping bands covering roughly 240–280 nm and 280–360 nm ranges. Both bands are assigned to the inter-configurational 4f⁷ → 4f⁶5d¹ transitions. In the spectrum recorded at 11 K, numerous vibronic features can be observed at the low-energy part of the spectrum. The PL spectrum at low-temperature (Fig. 3b) reveals a strong sharp line of the ⁶P_7/2 → ⁸S_7/2 (4f → 4f) transition and a well-resolved vibrational structure at its longer wavelengths/lower energy side. The PL spectrum changes significantly when the temperature increases. A broad PL band is observed at 300 K. It originates from the inter-configurational 4f⁶5d¹ → 4f⁷(⁸S_7/2) transition. Similarly, a significant change of the PL spectrum with temperature was observed in SrAl₁₂O₁₉:Eu²⁺.³⁰ The onset of the broadband is blue-shifted in respect to the main ⁶P_7/2 → ⁸S_7/2 line, indicating that the energetically lowest (emitting) 5d₁ level is located above the ⁶P_7/2 excited level of the 4f⁷ configuration of Eu²⁺, as discussed in the Introduction part and presented in Fig. 1b. The reason for such location of energy levels in the SrB₄O₇:Eu²⁺ material is the high coordination number (CN = 9) of the Eu²⁺/Sr²⁺, which makes the crystal field splitting of the 5d configuration fairly low.^7,9 Additionally, the low covalency of the bonds shifts the 4f⁶5d¹ configuration to a higher energy region. Solarz et al. proved a weak crystal field splitting for Sm²⁺ in the SrB₄O₇ host lattice, as well.³¹ The position of the 5d → 4f luminescence band of Eu²⁺ depends greatly on the host lattice properties. It may vary from 367 nm in SrB₄O₇:Eu²⁺ (Fig. 3b), up to even 850 nm in Ca₃Sc₂Si₃O₁₂:Eu²⁺.³²

Fig. 3 (a) PLE and PL spectra of SrB₄O₇:Eu²⁺ microparticles. The PLE spectrum was monitored at 367.2 nm and recorded at 11 and 300 K. (b) PL spectra recorded at 11 K and 300 K excited at 300 nm. In-depth analysis of the ZPL lines and vibronic transitions was presented in ref. 8.

For thermometric purposes, we measured the temperature dependence of the PL spectra (Fig. 4a) and decay traces of Eu²⁺ luminescence (Fig. 4b and c) in the T-range of 11–600 K. With the increasing temperature, three narrow lines, blue-shifted in respect to the ⁶P_7/2 → ⁸S_7/2 transition emerge (see Fig. 4a and Fig. S1 in ESI†). These three higher-energy components result from the thermally populated Stark sub-levels of the ⁶P_7/2 manifold,⁸i.e., from the ⁶P_7/2 → ⁸S_7/2 intra-configurational transition of Eu²⁺. Their spectral positions are in good agreement with the theoretically calculated splitting of the ⁶P_7/2 state into four Stark sub-levels, for the C_s symmetry.³³ With increasing temperature, the 4f⁶5d¹ → 4f⁷ broadband component suddenly appears and dominates the spectrum above 60 K. This obviously reflects the thermal population of the 5d₁ level, as discussed in the Introduction. Despite the increasing intensity of the inter-configurational emission, the ⁶P_7/2 → ⁸S_7/2 luminescence is clearly visible even up to ≈600 K, being potentially useful for luminescence thermometry applications.

Fig. 4 (a) Normalized (at 362.4 nm) temperature-dependent luminescence spectra of the SrB₄O₇:Eu²⁺ material, measured in the 11–600 K range upon excitation at 300 nm. Temperature-dependent decay traces of Eu²⁺ luminescence, recorded in the range (b) 11–90 K and (c) 100–600 K range.

Apart from the PL spectrum, the decay kinetics of Eu²⁺ emission also reveal a significant temperature dependence. Fig. 4b and c present decay traces of Eu²⁺ luminescence, recorded in the 11–600 K range. As the temperature increases, the decay time drastically shortens. Such characteristics of the luminescence kinetics are also in accordance with the energy scheme of Eu²⁺ presented in Fig. 1b. At the lowest temperature (11 K), the spin- and parity-forbidden ⁶P_7/2 → ⁸S_7/2 emission dominates, and this implicates the longest decay time. With increasing temperature, the contribution from the 5d → 4f parity-allowed (but spin-forbidden) transition systematically increases, shortening the decay time. Meijerink et al.⁸ have shown that the lowest level of the 4f⁶5d¹ excited configuration lies ∼130 cm⁻¹ above the lowest excited Stark sub-level of ⁶P_7/2 of Eu²⁺. Due to the mentioned small energy difference, efficient thermal coupling of the 4f⁶5d¹ and the 4f⁷ (⁶P_7/2) states can occur even at low temperatures. Note that both the intra-configurational ⁶P_7/2 → ⁸S_7/2 and the inter-configurational 4f⁶5d¹ → ⁸S_7/2 emission bands have the same terminal (ground) level, and the corresponding transitions start from the thermally coupled states.^7,34 In such a system, with the ΔE representing the energy difference between the involved levels, the rates of transitions from the levels being p_2→1 (the 4f → 4f in the discussed system) and p_3→1 (5d → 4f), the luminescence decay time (τ), at temperature T is well described by eqn (1):^34–37

The luminescence decay curves recorded at higher temperature values presented in Fig. 4c (having a dominant contribution of the 5d → 4f emission), exhibit nearly a single exponential behavior, although the perfect fits required the use of a two-exponential function, eqn (2):³⁸

where t₁ and t₂ are the decay time constants, τ_i are amplitudes of the relative components at time t = 0 and A₀ is the background intensity.

Thus, for temperature sensing applications, the average decay time, [small tau, Greek, macron] calculated according to eqn (3):³⁸

was used as a thermometric parameter.

The most potent changes of the luminescence decay time occur between ≈10 and 100 K, when it decreases from ∼300 μs to ∼14 μs, see Fig. 5a. At higher temperature values, the decay time further shortens monotonically, proving the growing contribution of the allowed 5d → 4f transition to the total emission of Eu²⁺.

Fig. 5 (a) Average decay time [small tau, Greek, macron] (eqn (3)) of SrB₄O₇:Eu²⁺ under 300 nm excitation and monitored at 362.4 nm. (b) Calculated relative thermal sensitivity of the investigated phosphor.

3.3. Luminescence decay time as a thermometric parameter

One of the most important parameters for characterizing and comparing the performance of a luminescence thermometer with other thermometers is its relative thermal sensitivity (S_r). Using the luminescence decay time as a thermometric parameter, S_r was calculated according to eqn (4):³⁹

Applying the luminescence decay time mode, the SrB₄O₇:Eu thermometer shows the maximal relative thermal sensitivity, S_m = 10.46%/K at 21 K. This is one of the highest relative thermal sensitivities reported, using the luminescence decay time as a thermometric parameter. At higher temperature values, the relative sensitivity decreases rapidly and monotonously. Yet, up to 600 K, the decay time may serve as a reliable thermometric parameter, which makes the operating range as broad as 11–600 K.

3.4. Luminescence intensity ratio as a thermometric parameter

The temperature-induced changes in the luminescence spectra are evidently attractive for luminescence thermometry, as well (see Fig. 4a). To determine the thermometric parameters, we chose four different spectral areas denoted as A1, A2, A3 (4f → 4f narrow emission lines), and A4 (5d → 4f broad emission band) presented in Fig. 6. It should be noted that the two emissions discussed above, i.e., 5d → 4f and the ⁶P_7/2 → ⁸S_7/2 (4f → 4f) superimpose strongly, which poses the question about the possibility of separating the corresponding, spectrally overlapping bands. Moreover, with increasing temperature, the number of narrow lines clearly contributes to the total luminescence changes, as the number of vibronic features and those related to the transitions from the thermally-populated Stark sub-levels change in the PL spectrum. This can be seen in Fig. 6a.

Fig. 6 (a) Temperature-dependent spectra of the SrB₄O₇:Eu²⁺ material in the T-range of 11–80 K and (b) the magnified part of the A1–A3 lines used for Δ parameter determination, (c) presents the spectra after subtracting the broadband fraction related to the 5d → 4f transition.

To calculate the intensities of the individual PL constituents, we separated the broad feature (band), corresponding to the 5d → 4f transition (A4, Fig. 6b), in order to extract the narrow-line components from the PL spectrum (A1–A3, Fig. 6c). For clarity and reproducibility, the applied procedure is depicted in Fig. S2a–c (ESI†), for representative temperature values. All the calculated values of the integrated areas are presented in Fig. S3 (ESI†). Consequently, the thermometric parameters Δ, were calculated for 5 different luminescence intensity ratios. Three of them (Δ₁, Δ₂, and Δ₃) involved the ratio between the broadband and one of the narrow lines. The next two fractions (Δ₄ and Δ₅) represent the intensity ratios between the narrow lines. Thus, the Δ parameters were calculated according to eqn (5):

The thermal evolution of the Δ₁₋₅ parameters is presented in Fig. 7a. The errors in the determination Δ₁₋₅ were calculated using eqn (6):⁴⁰

where I₁ and I₂ are the integrated intensities of two corresponding bands (A1–A5), and δI_i/I_i (i = 1, 2) values are estimated by calculating the signal-to-noise values.⁴⁰

Fig. 7 (a) Thermal evolution of Δ parameters for the SrB₄O₇:Eu²⁺ phosphor, (b) its relative thermal sensitivities, and (c) corresponding temperature uncertainties.

The temperature dependence of the thermometric parameters Δ_1–5 was successfully fitted using empirical functions, i.e., 3^th-order polynomials because the Boltzmann function did not give good fits. We will return to this observation in the Discussion section. Table S1 (ESI†) presents the fitting parameters for each of the Δ_1–5 thermometric parameters. The relative thermal sensitivities were calculated using eqn (3), where τ is replaced by Δ. Fig. 7b presents the calculated S_r values for each investigated Δ parameter. For S_r4, an unprecedentedly high value of the maximal relative sensitivity, S_m4 = 22.60% K⁻¹ at 16.5 K, was obtained. To the best of our knowledge, this is the best-reported value so far for inorganic materials doped with lanthanide ions. It should also be noted that for each thermometric parameter used, the calculated relative sensitivity is virtually always greater than 1% K⁻¹, within the entire operating T-range. Importantly, also for the other Δ parameters, the obtained S_m values are very high within the cryogenic T-range, (see Fig. 7b). It is worth noting that for Δ₁ and Δ₂ parameters the S_r values increase with temperature elevation, whereas for the Δ₃–Δ₅ ones the S_r values decrease, as clearly shown in Fig. 7b. Hence, utilizing the developed optical thermometer, the temperature of the system can be monitored in a very broad range, with high thermal sensitivities, better than 1% K⁻¹ over the entire operating T-range.

The high values of relative thermal sensitivities are accompanied with high thermal resolutions (reaching even 10⁻³ K in the cryogenic T-range), i.e., small uncertainty values (δT, Fig. 7c) calculated using eqn (7):⁴⁰

where δΔ/Δ stands for the error in the determination of Δ parameter (see eqn (6)). Fig. 7c presents the δT for all investigated Δ parameters. It is worth noting, that all δT values are in the 0.001–0.1 K range up to ≈100 K, which is very beneficial for various thermometric applications.

4. Discussion

It is notable that the maximal values of the relative thermal sensitivities (S_m) are very high both for the luminescence lifetimes (S_m ≈ 10.5% K⁻¹) and for the band intensity ratios (S_m ≈ 22.6% K⁻¹). The highest S_r values can be achieved in the T-range of 11–50 K. However, the operating range of the developed temperature sensor material – SrB₄O₇:Eu²⁺ – spans the 11–600 K range, using different sensing modes – lifetimes and intensity ratios. This makes the luminescent thermometer both a wide-range and extraordinarily sensitive tool. Its capability to work in different modes and to operate with multi-parameters makes it a universal sensor for thermometric purposes.

Table 1 compares the most important optical thermometry figure of merits, of the developed sensor SrB₄O₇:Eu²⁺ with other high-performance inorganic, luminescent thermometers reported in the literature. Recently, Marciniak et al.⁴¹ presented an original strategy to develop luminescence thermometers based on a reversible phase transition and achieved S_m = 11.8% K⁻¹ at 295 K (that is, around room temperature). In turn, the inorganic optical thermometer developed in this work is currently the leader in terms of thermal sensitivity in the cryogenic T-range.

Table 1 Thermometric performance of illustrative examples of the most sensitive, inorganic-based luminescence thermometers

Material	Measured parameter	S m (% K^−1)	T m (K)	T-range (K)	Equation type for Δ or τ-fitting	Ref.
a Eestimated based on figures in given references.
SrB4O7:Eu^2+	τ	10.47	21	11–600	Three-level system equation^36	This work
	Δ 1	4.21	280	50–280	Polynomial equation
	Δ 2	5.23	290	25–290
	Δ 3	18.65	11	11–150
	Δ 4	22.60	16.5	16.5–100
	Δ 5	12.76	25	25–110
LiYO2:Eu^3+	Δ	11.80	295	263–376^a	From experimental data	41
Sr2Ge0.75Si0.25O4:Pr^3+	Δ	9.20	65	18–600	Mott–Seitz equation	21
Sr2GeO4:Pr^3+	Δ	9.00	22	17–600	Mott–Seitz equation	20
YVO4:Nd^3+	Δ	9.00	123	123–873	Boltzmann equation	46
LaF3:Er^3+,Yb^3+	Δ	6.61	15	15–105	Mott–Seitz equation	47
Bi2MoO6: Er^3+,Tm^3+,Yb^3+	Δ	5.50	293	293–623	Polynomial equation	48
Y2Ti2O7:Pr^3+	Δ	5.25	289	289–573	Semi-empirical equation	49
SrTiO3:Lu^3+	Δ	4.60	200	120–290^a	From experimental data	50
ZnGa2O4:Cr^3+	Δ	2.80	310	77–450	Boltzmann equation	51
Bi2SiO5:Yb^3+,Tm^3+@SiO2	Δ	2.60	280	400–800	Boltzmann equation	52
TiO2:Sm^3+	Δτ	10.54	330	298–383^a	Not reported	53
		10.15	339
Al2O3:Sm^2+	Δτ	4.80	298	298–648	Boltzmann equation	54
		1.20	531	100–620
SrTiO3:Tb^3+	τ	8.83	180	10–230	From experimental data	50
SrB4O7:Sm^2+	τ	3.36	550	300–673	Boltzmann equation	55
Y2SiO5:Pr^3+	τ	2.67	417	11–500	Double-barrier equation	56
GAG:Mn^3+,Mn^4+	τ	2.08	249	123–350^a	Mott–Seitz equation	57

---

As we mentioned in the Introduction part, achieving a broad operating T-range combined with high relative sensitivity is a great challenge in the field of luminescence thermometry. In this paper, we showed that the Eu²⁺ luminescence in the strontium borate host matrix may fulfill the most important requirements of such thermometers. Thermal coupling of the excited states, the ⁶P_7/2 level of the 4f⁷ configuration, and the lowest level of the 4f⁶5d¹ configuration may offer both spectacular sensitivity and a wide operating range, spanning from cryogenic to at least 600 K. Since the energy of the ⁶P_7/2 level is basically fixed and independent of the host, the only way to design such thermometers is to maintain the energy of the 4f⁶5d¹ excited state slightly above the ⁶P_7/2 emitting level.

The energy of the lowest state of the 4f⁶5d¹ excited configuration is defined by the crystal-field splitting of this configuration, as well as by the ionicity of the chemical bonding of Eu²⁺. These bonds should be strongly ionic to shift the entire 4f⁶5d¹ configuration to higher energies. Then, the crystal-field splitting of the configuration should be as low as possible, to avoid shifting the lowest-energy 4f⁶5d¹ level too much downward. These requirements are in some sense contradictory since the ionic bonds exert a fairly high crystal-field strength. Yet, what may help is the introduction of the Eu²⁺ ion into a large site with a high coordination number of ligands that readily form ionic bonds. Apparently, these basic requirements are quite well-fulfilled in the SrB₄O₇:Eu phosphor. In order to shift the maximal relative sensitivity of the sensor to a higher temperature region, the gap between the lowest 4f⁶5d¹ level and the ⁶P_7/2 level below it should be greater than 130 cm⁻¹ in SrB₄O₇:Eu²⁺.

We would like to emphasize, yet, the difference (from the luminescence thermometry point of view) between the Eu²⁺ emission used here and the previously published properties of Pr³⁺ emission.^20–22 In thermometers using the latter dopant, emissions from the non-thermally-coupled levels were exploited and thermal quenching of such emissions was inherently utilized to get a high-performance sensor. This necessarily degraded the emission intensity deteriorating their thermal resolution (by decreasing the signal-to-noise ratio). On the contrary, in the case of the SrB₄O₇:Eu²⁺ sensor material, the emission of Eu²⁺ is not thermally quenched in the classical sense – the energy gained by the ion is not lost non-radiatively but is split between the two emitting levels of the dopant, instead. This is an important advantage, as the overall emission intensity is hardly dependent on temperature, with a respective reduction in the uncertainty of the measured temperature.

At first, it is surprising that the thermometric parameters based on luminescence intensity ratio (Fig. 7a) could not be effectively fitted using the Boltzmann function, despite the ⁶P_7/2 and the 4f⁶5d¹ emitting levels are obviously thermally coupled. In the earlier literature, one may find examples of similar divergence from anticipations.^42,43 Such a problem is typically encountered when luminescence intensity ratio in a wide range of temperature is analyzed or complex relaxation processes between (or from) the thermally coupled levels take place.^42,44 Such circumstances seem to exist in the SrB₄O₇:Eu²⁺ phosphor. The decay time of the 5d → 4f luminescence presents measurable (and useful for temperature sensing) changes from 11 to 600 K, while the 5d → 4f/4f → 4f luminescence intensity ratio appeared useful as thermometric parameter merely up to about 280 K. Detailed understanding of all aspects of the reasons of the non-Boltzmann dependence of the ratio of the emissions would require much more comprehensive examination of the phosphor. Possibly, more thorough computational exploration of deconvolution of the experimental spectra into the numerous components would be needed.

Finally, we note that the Eu²⁺ the characteristics of the luminescence of the SrB₄O₇:Eu²⁺ phosphor presents some similarity with emission of Cr³⁺ in some hosts. The latter ion shows interesting thermometric properties when its thermally coupled ⁴T₂ and ²E excited states are employed.^43,45 And also in this case this is the crystal field which plays an important role. When it is strong enough to lift the ⁴T₂ level above the ²E one, the Boltzmann distribution between them defines the thermometric performance of the selected phosphor. Yet, also for Cr³⁺, the Boltzmann law is obeyed over a limited range of temperature and in wide ranges it is not,⁴³ exactly as for Eu²⁺. Thus, the physics defining the SrB₄O₇:Eu²⁺ phosphor thermometric performance and standing behind the Cr³⁺ activated luminescence thermometers are fairly similar.

5. Conclusions

Here, we have shown that the selection of an appropriate host matrix (in our case SrB₄O₇) embedded with the desired, optically active divalent lanthanide ions (in this case Eu²⁺) is essential for the development of highly sensitive optical thermometers. The decisive role in designing such temperature sensors is an interplay between the inter- and intra-configurations transitions within the activator ion, where the relative energy difference of the corresponding excited states is crucial to achieving unprecedentedly high thermal sensitivity and great resolution of the final luminescent thermometer. In this work, it was found that the luminescence properties of the investigated SrB₄O₇:Eu²⁺ phosphor are very promising for thermometric purposes. Its emission spectra, as well as the decay kinetics, are strongly temperature-dependent. In the cryogenic temperature range, below ≈20 K, the emission results mainly from the intra-configurational ⁶P_7/2 → ⁸S_7/2 transition, while with increasing temperature, the broadband luminescence resulting from the inter-configurational 5d → 4f transition begins to appear, and increasingly contributes to the overall emission intensity. This results from the thermal coupling of the two emitting levels separated only by about 130 cm⁻¹. Another consequence of this thermal coupling between the excited ⁶P_7/2 and the lowest 4f⁶5d¹ levels is the strong temperature dependence of the luminescence decay time, which significantly shortens with increasing temperature. Altogether, the SrB₄O₇:Eu²⁺ phosphor presents very attractive properties for thermometric purposes. Its broad operating range in ratiometric mode spans the 11–280 K range, and its maximal relative sensitivities vary from ≈4 to above 22% K⁻¹ (depending on the Δ parameter used). Using the luminescence decay time as a thermometric parameter the operating range is even broader, namely 11–600 K, and the maximal relative thermal sensitivity was found to reach 10.46% K⁻¹, which is one of the highest values ever reported for relative sensitivity based on a luminescence decay time. Importantly, the high sensitivity values are in line with great thermal resolutions, which reach even 10⁻³ K in the cryogenic T-range. This paper proves that Eu²⁺ luminescence may offer much more for thermometry than previously thought.

Author contributions

The manuscript was written through the contributions of all authors. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the Polish National Science Centre (NCN) under grant #UMO 2018/29/B/ST5/00420. Publication partially financed by the program “Excellence Initiative – Research University”.

References

1. M. D. Dramićanin, J. Appl. Phys., 2020, 128, 040902.
2. C. D. S. Brites, S. Balabhadra and L. D. Carlos, Adv. Opt. Mater., 2019, 7, 1801239.
3. Y. Gao, Y. Cheng, T. Hu, Z. Ji, H. Lin, J. Xu and Y. Wang, J. Mater. Chem. C, 2018, 6, 11178–11183.
4. M. Runowski, P. Woźny, S. Lis, V. Lavín and I. R. Martín, Adv. Mater. Technol., 2020, 1901091.
5. M. A. Hernández-Rodríguez, K. Kamada, A. Yoshikawa, J. E. Muñoz-Santiuste, A. Casasnovas-Melián, I. R. Martín, U. R. Rodríguez-Mendoza and V. Lavín, J. Alloys Compd., 2021, 161188.
6. L. Marciniak, K. Elzbieciak-Piecka, K. Kniec and A. Bednarkiewicz, Chem. Eng. J., 2020, 388, 124347.
7. G. Blasse and B. Grabbmaier, Luminescent Materials, Springer-Verlag, Berlin Heidelberg, 1994.
8. A. Meijerink, J. Nuyten and G. Blasse, J. Lumin., 1989, 44, 19–31.
9. P. Dorenbos, J. Lumin., 2013, 135, 93–104.
10. J. G. Bunzli and S. V. Eliseeva, Springer Ser. Fluoresc., 2011, 1–45.
11. M. Suta and C. Wickleder, J. Lumin., 2019, 210, 210–238.
12. M. Suta, F. Lavoie-Cardinal and C. Wickleder, Angew. Chem., Int. Ed., 2020, 59, 10949–10954.
13. F. Datchi, R. LeToullec and P. Loubeyre, J. Appl. Phys., 1997, 81, 3333–3339.
14. T. Zheng, M. Runowski, P. Woźny, S. Lis and V. Lavín, J. Mater. Chem. C, 2020, 8, 4810–4817.
15. G. Blasse, G. J. Dirksen and A. Meijerink, Chem. Phys. Lett., 1990, 167, 41–44.
16. P. Solarz, J. Komar, M. Głowacki, M. Berkowski and W. Ryba-Romanowski, RSC Adv., 2017, 7, 21085–21092.
17. W. Xu and J. R. Peterson, J. Alloys Compd., 1997, 249, 213–216.
18. M. De Jong, A. Meijerink, R. A. Gordon, Z. Barandiarán and L. Seijo, J. Phys. Chem. C, 2014, 118, 9696–9705.
19. M. De Jong, A. Meijerink, Z. Barandiarán and L. Seijo, J. Phys. Chem. C, 2014, 118, 17932–17939.
20. C. D. S. Brites, K. Fiaczyk, J. F. C. B. Ramalho, M. Sójka, L. D. Carlos and E. Zych, Adv. Opt. Mater., 2018, 1701318.
21. M. Sójka, J. F. C. B. Ramalho, C. D. S. Brites, K. Fiaczyk, L. D. Carlos and E. Zych, Adv. Opt. Mater., 2019, 7, 1901102.
22. M. Sójka, C. D. S. Brites, L. D. Carlos and E. Zych, J. Mater. Chem. C, 2020, 8, 10086–10097.
23. P. Bolek, J. Zeler, C. D. S. Brites, J. Trojan-Piegza, L. D. Carlos and E. Zych, Chem. Eng. J., 2021, 421, 129764.
24. T. Zheng, M. Sójka, M. Runowski, P. Woźny, S. Lis and E. Zych, Adv. Opt. Mater., 2021, 9, 2101507.
25. Y. Pan, X. Xie, Q. Huang, C. Gao, Y. Wang, L. Wang, B. Yang, H. Su, L. Huang and W. Huang, Adv. Mater., 2018, 30, 3–8.
26. L. Zhou, P. Du and L. Li, Sci. Rep., 2020, 10, 1–11.
27. D. Stefańska, M. Stefanski and P. J. Dereń, J. Alloys Compd., 2021, 863, 1–7.
28. R. Zhou, C. Liu, L. Lin, Y. Huang and H. Liang, Chem. Eng. J., 2019, 369, 376–385.
29. R. Zhou, L. Lin, H. Zhao, T. Deng and J. Li, Mater. Chem. Front., 2021, 5, 6071–6081.
30. D. Dutczak, T. Jüstel, C. Ronda and A. Meijerink, Phys. Chem. Chem. Phys., 2015, 17, 15236–15249.
31. P. Solarz, M. Karbowiak, M. Głowacki, M. Berkowski, R. Diduszko and W. Ryba-Romanowski, J. Alloys Compd., 2016, 661, 419–427.
32. I. V. Berezovskaya, V. P. Dotsenko, A. S. Voloshinovskii and S. S. Smola, Chem. Phys. Lett., 2013, 585, 11–14.
33. J. G. Bunzli and S. V. Eliseeva, Springer Ser. Fluoresc., 2011, 1–45.
34. K. Fiaczyk, A. J. Wojtowicz and E. Zych, J. Phys. Chem. C, 2015, 119, 5026–5032.
35. B. Di Bartolo and R. Summitt, Phys. Today, 1970, 23, 105–107.
36. G. A. M. Dalhoeven and G. Blasse, J. Solid State Chem., 1981, 39, 35–38.
37. P. Kisliuk and C. A. Moore, Phys. Rev., 1967, 160, 307–312.
38. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, US, Boston, MA, 2006, pp. 1–26.
39. C. D. S. Brites, P. P. Lima, N. J. O. Silva, A. Millán, V. S. Amaral, F. Palacio and L. D. Carlos, Nanoscale, 2012, 4, 4799.
40. C. D. S. Brites, A. Millán and L. D. Carlos, Handbook on the Physics and Chemistry of Rare Earths, 2016, vol. 49, pp. 339–427.
41. L. Marciniak, W. Piotrowski, M. Szalkowski, V. Kinzhybalo, M. Drozd, M. Dramicanin and A. Bednarkiewicz, Chem. Eng. J., 2022, 427, 131941.
42. R. G. Geitenbeek, H. W. De Wijn and A. Meijerink, Phys. Rev. Appl., 2018, 10, 1.
43. M. Back, J. Ueda, M. G. Brik and S. Tanabe, ACS Appl. Mater. Interfaces, 2020, 12, 38325–38332.
44. M. Suta and A. Meijerink, Adv. Theory Simul., 2020, 3, 1–32.
45. M. Back, J. Ueda, H. Nambu, M. Fujita, A. Yamamoto, H. Yoshida, H. Tanaka, M. G. Brik and S. Tanabe, Adv. Opt. Mater., 2021, 9, 1–11.
46. A. A. Kalinichev, M. A. Kurochkin, E. V. Golyeva, A. V. Kurochkin, E. Lähderanta, M. D. Mikhailov and I. E. Kolesnikov, J. Lumin., 2018, 195, 61–66.
47. A. M. Kaczmarek, M. K. Kaczmarek and R. Van Deun, Nanoscale, 2019, 11, 833–837.
48. T. Zheng, M. Runowski, N. Stopikowska, M. Skwierczyńska, S. Lis, P. Du and L. Luo, J. Alloys Compd., 2021, 890, 161830.
49. R. Lei, X. Luo, Z. Yuan, H. Wang, F. Huang, D. Deng and S. Xu, J. Lumin., 2019, 205, 440–445.
50. W. Piotrowski, M. Kuchowicz, M. Dramićanin and L. Marciniak, Chem. Eng. J., 2022, 428, 131165.
51. J. Ueda, M. Back, M. G. Brik, Y. Zhuang, M. Grinberg and S. Tanabe, Opt. Mater., 2018, 85, 510–516.
52. E. Casagrande, M. Back, D. Cristofori, D. Cristofori, J. Ueda, S. Tanabe, S. Palazzolo, F. Rizzolio, F. Rizzolio, V. Canzonieri, V. Canzonieri, E. Trave and P. Riello, J. Mater. Chem. C, 2020, 8, 7828–7836.
53. M. D. Dramićanin, Ž. Antić, S. Ćulubrk, S. P. Ahrenkiel and J. M. Nedeljković, Nanotechnology, 2014, 25, 485501.
54. A. Ćirić, S. Stojadinović, Z. Ristić, I. Zeković, S. Kuzman, Ž. Antić and M. D. Dramićanin, Adv. Mater. Technol., 2021, 6, 1–10.
55. Z. Cao, X. Wei, L. Zhao, Y. Chen and M. Yin, ACS Appl. Mater. Interfaces, 2016, 8, 34546–34551.
56. M. Sójka, M. Runowski, P. Woźny, L. D. Carlos, E. Zych and S. Lis, J. Mater. Chem. C, 2021, 9, 13818–13831.
57. L. Marciniak and K. Trejgis, J. Mater. Chem. C, 2018, 6, 7092–7100.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1tc05022b

---