Optical temperature sensing of rare-earth ion doped phosphors

Abstract

Accurate and reliable temperature measurement of many special inaccessible objects is a challenging task. Optical temperature sensing is a promising method to achieve it. The current status of optical thermometry of rare-earth ion doped phosphors is reviewed in detail. Based on the mechanisms of optical temperature sensing of different phosphors, temperature dependent luminescence spectra, the fluorescence intensity ratio technique in the data fitting process, and errors of the energy difference between thermally coupled levels, we describe the recent developments in the use of optical thermometry materials. The most important results obtained in each case are summarized, and the main challenges that we need to overcome are discussed. Research in the field of phosphor sensors has shown that they have significant advantages compared to conventional sensors in terms of their properties like greater sensitivity, freedom from electromagnetic interference, long path monitoring, and independence of compatibility with electronic devices.

---

Xiangfu Wang

Xiangfu Wang received his PhD degree from Nanjing University of Aeronautics and Astronautics, in 2012. From 2012 to present, he was promoted to an associate professor in the College of Electronic Science and Engineering at Nanjing University of Posts and Telecommunications. His current research concentrates on the synthesis, growth mechanism, photoluminescence and optical temperature sensing of rare-earth ion doped nanomaterials and transparent glass ceramics.

Qing Liu

Qing Liu obtained her Bachelor degree from Nanjing University of Posts and Telecommunications, in 2015. From 2015 to present, she works for a Master degree in Xiangfu Wang’s group. Her current research concentrates on the photoluminescence of rare-earth ion doped nanomaterials.

Yanyan Bu

Yanyan Bu obtained her Master degree from Xiangtan University, in 2010. From 2010 to present, she works in Nanjing University of Posts and Telecommunications. From 2014 to present, she works for a Doctor degree in Xiaohong Yan’s group. Her current research concentrates on optical temperature sensing of rare-earth ion doped transparent glass ceramics.

Chun-Sheng Liu

Chun-Sheng Liu received his PhD in 2011 from Institute of Solid State Physics, Chinese Academy of Sciences. Then he was awarded an AvH (Alexander von Humboldt Foundation) Fellowship (2011–2013) in Technische Universitaet München, Germany. In 2014, he was promoted to a full professor in the College of Electronic Science and Engineering at Nanjing University of Posts and Telecommunications (NJUPT). He was elected as Specially-Appointed Professors by NJUPT in Jiangsu Province. His main research fields are functional materials for hydrogen storage, gas sensors, and photocatalysis.

Tao Liu

Tao Liu obtained his Bachelor degree from Nanjing University of Posts and Telecommunications (NJUPT), China, in 2012, and he received his M.S degree from NJUPT in 2015 under the supervision of Prof. Xiaohong Yan and Prof. Xiangfu Wang. His current research concentrates on Computer Graphics.

Xiaohong Yan

Xiaohong Yan received his PhD degree from Institute of Solid State Physics, Chinese Academy of Science, in 1997. From 2010 to present, he works as a full professor in the College of Electronic Science and Engineering at Nanjing University of Posts and Telecommunications. His current research concentrates on transport and manipulation of quantum systems and new energy materials.

---

1. Introduction

Temperature is a key parameter to induce some changes of cellular events, physical and chemical properties of functional nanomaterials, functional degradation of microcircuits, and so on. Contact temperature measurement, just like a thermometer, is an ideal method to study some phenomena that change with temperature. However, especially at the sub-micron scale, for temperature fluctuations of microcircuits and intracellular liquids, the conventional thermometer is not able to make measurements.^1–3 Brites and Jaque reported some examples of luminescent and non-luminescent thermometers working at the nanometric scale.^4,5 Even so, it is difficult to measure directly the temperatures of inaccessible objects, such as high-voltage power stations, coal mines, and volcanic and corrosive circumstances.^6–9 Therefore, the ability to glean the temperature of the above objects could have valuable repercussions leading to novel insight about their properties induced by a temperature change.

Optical temperature sensing, in contrast to other thermometers, is a promising method to achieve contactless measurement and large-scale imaging. Optical temperature sensing is based on monitoring the emission intensity change of luminescent materials induced by temperature when they interact with physical systems. Phosphors with an intense emission intensity are used as temperature detectors to convert the measured spectrum to temperature. It is easy to calibrate and calculate the temperature change range only through the main parameters of luminescence including intensity, effective bandwidth, spectrum shape, spectral shift and lifetime.

Recently, CdSe and CdTe semiconducting quantum dots were used as optical temperature nanoprobes in the low temperature range, since the peak position of the emission wavelength changes as a function of temperature.^10,11 Organic compounds, such as Rhodamine B, the Ru–phen complex, DPTB dissolved in MOE, and so on, were used for dye-based intensity luminescence nanothermometry, due to their the spectral properties depending on many factors, such as the solvent, concentration, pH and temperature.^5,11,12 These quantum dot nanocrystals and organic compounds are easily oxidized at high temperature, and are only available in the low temperature range. To overcome this, rare-earth ion doped oxide crystals, glass, core–shell heterojunction nanoparticles, and transparent glass ceramics containing fluoride nanocrystals were fabricated and used as optical temperature sensors. The present review aims to describe the latest progress of optical thermometry based on the luminescence of rare-earth ion doped phosphors. It contains fundamental principles, different rare-earth ion doped phosphors for optical thermometry, a conclusion and the main challenges.

2. Fundamental principles of optical temperature sensing

Phosphor thermometry is a non-contact technique that uses luminescence signals to measure temperature remotely. As shown in Fig. 1, phosphors are composed of a host and some luminescent ions, and will emit visible, infrared, or ultraviolet radiation upon excitation from an external light source. The intensity, wavelength, and lifetime of the emission bands are used to determine the temperature of a surface. Thus, the choice of luminescent ions is important to measure the temperature change around phosphors. Among the metal ions used as luminescence centers, the trivalent rare-earth ions are especially attractive, due to abundant energy levels located at a wide wavelength range from ultraviolet to infrared.¹³ Possessing real intermediate energy levels, trivalent rare-earth ions can give out desired emissions via various energies, due to abundant energy levels of 4f configurations. Inheriting their native intra-configurational transitions, trivalent rare-earth ion activated luminescent materials have received consistent attention due to large Stokes/anti-Stokes shifts, long luminescence lifetimes, and sharp band emissions.¹⁴ Additionally, some trivalent rare-earth ions own a couple of adjacent levels with a very small energy gap (ΔE) of about 100–2000 cm⁻¹, such as Er³⁺: ²H_11/2 and ⁴S_3/2; Tm³⁺: ³F_2,3 and ³H₄; Ho³⁺: ⁵S₂ and ⁵F₄; Nd³⁺: ⁴F_5/2 and ⁴F_3/2; Dy³⁺: ⁴I_15/2 and ⁴F_9/2; Eu³⁺: ⁵D₁ and ⁵D₀, and so on. As shown in Fig. 1, in the photoluminescence process, these adjacent energy levels, the upper level and lower level, can be thermally populated and depopulated through changing the environmental temperature around the phosphors. The two adjacent energy levels were called thermally coupled energy levels (TCL). The luminescence intensity ratio between I_U and I_L will change regularly with the temperature increase. A function relation between the luminescence intensity ratio and temperature can be determined through fitting some data points at different temperatures.

Fig. 1 Schematic illustration of the basic mechanism of the optical thermometry process in rare-earth ion doped phosphors.

The luminescence intensity of an emission band can be expressed as:

I_ij = hvA_ijN_i (1)

where hv is the transition energy per photon from an i state to a j state, A_ij is the spontaneous radiative emission probability, and N_i is the state population of the i state.¹⁵

The term v is expressed as:

where λ is the mean wavelength of the transition, his Planck’s constant, and c is the speed of light.

In conclusion, I_ij can be expressed as:

The ratio of the luminescence from each thermally coupled level of active ions is modified as:

where I_U and I_L are the fluorescence intensities generated by the radiative transitions from the upper and lower thermally coupled levels to the ground level.

Thus, the term R can be given as the following:

where λ_L is the mean wavelength of the L → j transition, λ_U is the mean wavelength of the U → j transition, A_Uj is the spontaneous radiative emission probability of the U → j transition, and A_Lj is the spontaneous radiative emission probability of the L → j transition. N_U and N_L are the state populations of the upper and lower levels.

The terms A_Uj, A_Lj, N_U, and N_L are dependent on the temperature. The population process of the N_U and N_L levels obeys the Boltzmann distributing law.¹⁶ Thus, R can be expressed as:

where A is a fitting constant that depends on the experimental system and intrinsic spectroscopic parameters; ΔE_f is the fitting energy difference between thermally coupled levels; K is the Boltzmann constant (298 K, KT ≈ 210 cm⁻¹); T is the absolute temperature; and B is an offset.^17,18 The term B includes the overlapping of fluorescence peaks originating from the two individual thermally coupled energy levels, and stray light originating from other energy levels or from the excitation. B is expressed as:where n defines the fraction of the total fluorescence intensity, and m defines the fraction of the total intensity.

The sensitivity is a key parameter to value the possibility of practical applications. The sensitivity of optical thermometry is the rate of change of R in response to the variation of temperature.^17,19,20 To allow comparison between the sensitivities obtained from different thermally coupled levels, the relative sensitivity S_R and the absolute sensitivity S_A are defined as:

and where ΔE_f is the fitting energy difference between thermally coupled levels.

The terms S_R and S_A are dependent on ΔE_f. If ΔE_f agrees well with the experimental energy difference ΔE_m, the values of S_R and S_A are accurate. The error δ between ΔE_f and ΔE_m is expressed as:

where ΔE_m is calculated from the two analyzed peaks in the spectrum. The error δ is a key parameter to determine whether the value of ΔE_f agrees well with the experimental value of ΔE_m. A large value of δ means that the energy transfer between thermally coupled levels and other levels is not neglected, and the population of thermally coupled levels at high temperature is induced by the routes of the Boltzmann distribution and energy transfer.^4,14,17

3. Rare-earth ion doped phosphors for optical thermometry

Phosphors are composed of inorganic oxides and ceramic materials which means that phosphors are resistant to oxidation in high temperature environments and are non-reactive with harsh chemicals. To obtain excellent optical temperature sensors, trivalent lanthanide ion doped phosphors were synthesized widely.^21–80 At present, trivalent lanthanide ions, such as Er³⁺, Tm³⁺, Ho³⁺, Nd³⁺, Dy³⁺, and Eu³⁺, have been used as luminescence centers (or activators) in the process of optical thermometry, as shown in Fig. 2. Host materials used for optical thermometry involve nanorods,⁶⁰ spherical and tetragonal nanoparticles,⁵⁹ core–shell particles,⁶¹ hollow nanoparticles, glass ceramics containing fluoride nanocrystals,^23,68,71 fibers,¹⁷ and oxide bulks.^{34,39–54,58} Phosphors are excited with a light source, and the emitted luminescence can be in the ultraviolet, visible, or even in the infrared region.

Fig. 2 Different types and luminescence transitions of rare-earth ion doped phosphors used for optical thermometry. Data are summarized according to ref. 21–80.

For optical temperature sensors, the TCL should satisfy some conditions that depend strongly on the host matrix into which the rare-earth ions are doped. The factors affecting the TCL of the rare-earth ions are as follows:

1. The separation between the TCL should be more than 200 cm⁻¹ to avoid strong overlapping of the two fluorescence wavelengths, and be less than 2000 cm⁻¹ to avoid too small a population in the upper level for the temperature range of interest.

2. In order to achieve a sufficient fluorescence intensity from the upper level transition, the radiative transitions from the upper level should dominate its non-radiative transitions.

TCL proven by experiment are listed as follows: Er³⁺: ²H_11/2/⁴S_3/2, and ⁴D_7/2/⁴G_9/2; Tm³⁺: ³F_2,3/³H₄, and ¹G_4(a)/¹G_4(b); Ho³⁺: ⁵S₂/⁵F₄, ⁵F_2,3/³K₈, and ⁵G₆/⁵F₁; Nd³⁺: ⁴F_5/2/⁴F_3/2, ⁴F_7/2/⁴F_3/2, and ⁴F_7/2/⁴F_5/2; Dy³⁺: ⁴I_5/2 and ⁴F_9/2; and Eu³⁺: ⁵D₁/⁵D₀. The TCL are populated by the up-conversion and down-conversion processes of the above trivalent rare-earth ions. Optical thermometry has been achieved by analyzing the temperature dependent luminescence properties originating from the transitions from the TCL to the other levels.

3.1 Optical thermometry based on Er³⁺ doped phosphors

Trivalent erbium Er³⁺ has a 4f¹¹ electronic configuration and dense energy levels located at a wide wavelength range from ultraviolet to infrared, which is suitable to absorb and emit ultraviolet, visible and infrared luminescence. The Er³⁺ ion has two couples of adjacent TCL, ²H_11/2 and ⁴S_3/2, and ⁴D_7/2 and ⁴G_9/2, whose relative emission intensities are strongly temperature dependent. Among the different rare-earth ions capable of single-center phosphors, Er³⁺ is probably the most used one, due to its very intense two green emission bands originating from the ²H_11/2 and ⁴S_3/2 TCL. As shown in Fig. 3, the mechanism of optical thermometry based on Er³⁺ doped phosphors can be summarized as follows:

Fig. 3 The mechanism of optical thermal sensing through (a) Er³⁺ down-conversion under ultraviolet (blue) excitation, (b) Er³⁺ up-conversion under infrared excitation, and (c) Yb³⁺–Er³⁺ energy transfer under infrared excitation.

(1) Down-conversion of Er³⁺ ions under ultraviolet (or blue) excitation:^21–26 under ultraviolet (or blue) excitation, Er³⁺ ions are excited directly to the higher excited states by ground state absorption (GSA), as shown in Fig. 3(a). Some ions in the excited states relax to the next lower energy levels ²H_11/2 and ⁴S_3/2 through the process of non-radiative relaxation, making the ²H_11/2 and ⁴S_3/2 levels populated. Er³⁺ ions in the ²H_11/2 and ⁴S_3/2 levels radiatively relax to the ⁴I_15/2 level, giving two green emissions with the intensity of I_U and I_L. Er³⁺ ions in the ²H_11/2 and ⁴S_3/2 levels radiatively relax to the ⁴I_13/2 level, giving two green emission bands with the intensity of I^′_U and I^′_L. The luminescence intensity ratios, I_U/I_L and I^′_U/I^′_L, change with the temperature of the phosphors. These temperature dependent fluorescence intensity ratios are used as the precise evaluation scale of optical temperature sensing.

Recently, based on down-conversion luminescence emissions centered at 524 nm (²H_11/2 → ⁴I_15/2) and 547 nm (⁴S_3/2 → ⁴I_15/2) of Er³⁺ ions, as shown in Fig. 4, optical temperature sensing in the range from 313 K to 713 K was studied by Sui et al. through analyzing the temperature dependent term R in Er³⁺ doped Te–Ge–Zn–Na glass under 379 nm excitation.²¹ A maximum sensitivity of 0.0085 K⁻¹ at 596 K was obtained in the Te₄₀Ge₄₀Zn_9.5Na₁₀ glass. Under 406 nm excitation, the intensity ratio of green emissions at 523 nm and 545 nm was studied by González et al. in Er³⁺ doped fluoroindate glass with a maximum sensitivity of 0.0028 K⁻¹ for 425 K.²² Under 488 nm excitation, the thermalized levels ⁴S_3/2 and ²H_11/2 were studied in Er³⁺ doped NaYF₄ nanocrystalline glass ceramic with a maximum sensitivity of 66 × 10⁻⁴ K⁻¹ for 570 K.²³ Similarly, the maximum sensitivity of 79 × 10⁻⁴ K⁻¹ for 630 K was obtained in Er³⁺ doped zinc fluorophosphate glass, and 79 × 10⁻⁴ K⁻¹ for 541 K was obtained in Er³⁺ doped fluorotellurite glass.^24,25 Under 532 nm excitation, González et al. demonstrated novel optical temperature thermometry based on infrared emissions centered at 800 nm (²H_11/2 → ⁴I_13/2) and 850 nm (⁴S_3/2 → ⁴I_13/2) in Er³⁺ doped Sr–Ba–Nb–B glass ceramic.²⁶ A maximum sensitivity of 0.0017 K⁻¹ for 600 K was obtained. The hosts, excitation wavelength (λ_ex), emission wavelength (λ_em), the involved transitions, S_R and S_A values, and the temperature range for phosphors doped with Er³⁺ ions are summarized in Table 1.

Fig. 4 Temperature dependent Er³⁺ green emission spectra of 0.5 mol% Er³⁺ doped (a) Te₈₀Zn_9.5Na₁₀ (TG08) glass, (b) Te₄₀Ge₄₀Zn_9.5Na₁₀ (TG44) glass, (c) Ge₈₀Zn_9.5Na₁₀ (TG80) glass, and (d) temperature dependent R values of the two green emissions of Er³⁺. Reprinted from ref. 21 with permission of Springer.

Table 1 The maximum sensitivity values of Er³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Er^3+	Tellurite glass	379	524, 547	^2H11/2, ^4S3/2 → ^4I15/2	313–713	0.0085 K^−1 (596 K)	1108.3/T^2	21
Er^3+	In–Zn–Sr–Ba glass	406	523, 545	^2H11/2, ^4S3/2 → ^4I15/2	125–425	0.0028 K^−1 (425 K)	1167.6/T^2	22
Er^3+	α-NaYF4 glass ceramics	488	525, 550	^2H11/2, ^4S3/2 → ^4I15/2	300–720	0.0024 K^−1 (540 K)	1051.51/T^2	23
Er^3+	PKAZLF glass	488	524, 550	^2H11/2, ^4S3/2 → ^4I15/2	298–773	0.0079 K^−1 (630 K)	1238.8/T^2	24
Er^3+	Te–Pb–Al glass	488	530, 550	^2H11/2, ^4S3/2 → ^4I15/2	100–573	0.0079 K^−1 (541 K)	1149.43/T^2	25
Er^3+	Sr–Ba–Nb–B glass	532	800, 850	^2H11/2, ^4S3/2 → ^4I13/2	300–700	0.0017 K^−1 (600 K)	1255.14/T^2	26
Er^3+	Te–Pb–Al glass	800	530, 550	^2H11/2, ^4S3/2 → ^4I15/2	300–550	0.0054 K^−1 (540 K)	1085.57/T^2	27
Er^3+	ZBLALiP glass	805	522, 546	^2H11/2, ^4S3/2 → ^4I15/2	150–850	0.0023 K^−1 (495 K)	981/T^2	28
Er^3+	Ba(Zr,Ca)TiO3 ceramics	980	525, 550	^2H11/2, ^4S3/2 → ^4I15/2	200–443	0.0044 K^−1 (443 K)	1135.5/T^2	29
Er^3+	Si–B–Ba–Na glass	978	534, 549	^2H11/2, ^4S3/2 → ^4I15/2	296–673	0.0023 K^−1 (296 K)	335/T^2	30
Er^3+	BaTiO3 nanocrystals	980	526, 547	^2H11/2, ^4S3/2 → ^4I15/2	322–466	1200 K^−1	940/T^2	31
Er^3+	PLZT ceramics	980	534, 565	^2H11/2, ^4S3/2 → ^4I15/2	310–883	0.004 °C^−1 (610 °C)	1096.92/T^2	32
Er^3+, Mo^6+	YbAG	976	522, 546	^2H11/2, ^4S3/2 → ^4I15/2	295–973	0.0048 K^−1 (467 K)	900.8/T^2	33
Er^3+	ZnO	978	536, 553	^2H11/2, ^4S3/2 → ^4I15/2	353–973	0.0062 °C^−1 (170 °C)	880.11/T^2	34
Er^3+	Chalcogenide glass	1540	530, 555	^2H11/2, ^4S3/2 → ^4I15/2	293–493	0.0102 °C^−1	1135.23/T^2	35
Er^3+	Na0.82Ca0.08Er0.16Y0.853F4	1540	523, 542	^2H11/2, ^4S3/2 → ^4I15/2	5–300	0.0022 K^−1 (338 K)	958.28/T^2	36
Er^3+, Yb^3+	α-NaYF4	980	525, 545	^2H11/2, ^4S3/2 → ^4I15/2	298–318	0.0030 K^−1 (515 K)	1028/T^2	37
Er^3+, Yb^3+	β-NaLuF4	980	256, 276	^4D7/2, ^4G9/2 → ^4I15/2	303–523	0.0052 K^−1 (303 K)	384/T^2	38
Er^3+, Yb^3+, Eu^3+	Y2O3	980	523, 551	^2H11/2, ^4S3/2 → ^4I15/2	303–593	0.0103 K^−1 (593 K)	1062.50/T^2	39
Er^3+, Yb^3+	Y2O3	978	539, 564	^2H11/2, ^4S3/2 → ^4I15/2	150–300	0.0528 K^−1 (150 K)	1173.39/T^2	40
Er^3+, Yb^3+	Y2O3	980	528, 556	^2H11/2, ^4S3/2 → ^4I15/2	93–613	0.0044 K^−1 (427 K)	886.08/T^2	41
Er^3+, Yb^3+, Eu^3+	Y2O3	980	523, 551	^2H11/2, ^4S3/2 → ^4I15/2	301–403	0.0008 K^−1 (327 K)	682.6/T^2	42
Er^3+, Yb^3+	Al2O3	978	523, 545	^2H11/2, ^4S3/2 → ^4I15/2	295–973	0.0051 K^−1 (495 K)	964.1/T^2	43
Er^3+, Yb^3+	BaMoO4	980	531, 552	^2H11/2, ^4S3/2 → ^4I15/2	303–523	0.0206 K^−1 (463 K)	873.38/T^2	44
Er^3+, Yb^3+	Y2SiO5 pulsed	975	530, 550	^2H11/2, ^4S3/2 → ^4I15/2	300–600	0.0070 K^−1 (600 K)	1226/T^2	45
Er^3+, Yb^3+, Nd^3+	Y2SiO5	808	528, 556	^2H11/2, ^4S3/2 → ^4I15/2	298–753	0.00095 K^−1(439 K)	710.87/T^2	46
Er^3+, Yb^3+	SrWO4	980	525, 547	^2H11/2, ^4S3/2 → ^4I15/2	300–518	0.01498 K^−1 (403 K)	866.17/T^2	47
Er^3+, Yb^3+	YVO4	980	524, 554	^2H11/2, ^4S3/2 → ^4I15/2	300–485	0.01169 K^−1 (380 K)	774.1/T^2	48
Er^3+, Yb^3+	CaWO4	980	384, 408	^4G11/2, ^2H9/2 → ^4I15/2	303–873	0.0073 K^−1 (873 K)	2109.31/T^2	51
Er^3+, Mo^6+	Yb2Ti2O7	976	532, 546	^2H11/2, ^4S3/2 → ^4I15/2	290–610	0.0074 K^−1 (340 K)	679.2/T^2	52
Er^3+, Yb^3+	LiNbO3	980	530, 550	^2H11/2, ^4S3/2 → ^4I15/2	285–453	0.0075 K^−1 (310 K)	1250/T^2	54
Er^3+, Yb^3+	Na0.5Bi0.5TiO3 ceramics	980	525, 550	^2H11/2, ^4S3/2 → ^4I15/2	173–553	0.0035 K^−1 (493 K)	1017.12/T^2	49
Er^3+, Yb^3+	NaBiTiO3 ceramics	980	525, 550	^2H11/2, ^4S3/2 → ^4I15/2	163–613	0.0031 K^−1 (400 K)	827.26/T^2	50
Er^3+, Yb^3+	Bi7Ti4NbO21	980	547, 670	^4S3/2, ^4F9/2 → ^4I15/2	153–553	0.0044 K^−1		53
Er^3+, Yb^3+	TeO2–WO3 glass	980	527, 551	^2H11/2, ^4S3/2 → ^4I15/2	300–690	0.0029 K^−1 (690 K)	976.75/T^2	55

---

(2) Up-conversion of Er³⁺ ions under infrared excitation: under infrared excitation, the excited states of Er³⁺ ions are populated by the GSA and successive excited state absorption (ESA),^27–36 as shown in Fig. 3(b). It needs three 1540 nm infrared photons and two 800 nm (or 980 nm) infrared photons to populate the thermally coupled ²H_11/2 and ⁴S_3/2 energy levels. The energy transfer between Er³⁺ ions is thought of as the main up-conversion mechanism, only when the Er³⁺ concentration of the optically active center is high enough.¹³ The ions in the high energy excited states relax to the next lower energy levels ²H_11/2 and ⁴S_3/2 through non-radiative relaxation, giving two green emissions with the intensities of I_U and I_L. The temperature dependent fluorescence intensity ratio, I_U/I_L, is used as the precise evaluation scale of optical temperature sensing.

Based on the up-conversion luminescence emissions centered at 524 nm (²H_11/2 → ⁴I_15/2) and 547 nm (⁴S_3/2 → ⁴I_15/2) of Er³⁺ ions, the optical temperature thermometry was studied by Luis et al. in a fluorotellurite glass under 800 nm excitation.²⁷ A better behaviour as a temperature sensor has been obtained for the less Er³⁺ concentrated glass with a maximum sensitivity of 54 × 10⁻⁴ K⁻¹ at 540 K. Based on the thermalization effects between the upper levels responsible for green fluorescence at 522 nm and 546 nm under 805 nm laser excitation, new optical temperature sensing using the micrometer sized ZBLALiP spherical cavity was explored by Cai et al.²⁸ Using a diode laser emitting at about 980 nm as the excitation source, the optical temperature sensing properties based on green emissions of Er³⁺ ions were studies in BZT-BCT ferroelectric ceramics, silicate glass, BaTiO₃ nanocrystals, PLZT transparent ceramics, Yb₃Al₅O₁₂ nanocrystals, and ZnO nanocrystals, respectively.^29–34 The maximum sensitivity of 48 × 10⁻⁴ K⁻¹ for 467 K was obtained in Er³⁺ doped Yb₃Al₅O₁₂ nanocrystals. Under 1540 nm infrared excitation, based on green luminescence emissions of Er³⁺ ions, the optical temperature sensing properties were studied in Ga₂S₃–La₂O₃ chalcogenide glass, and the Na_0.82Ca_0.08Er_0.16Y_0.853F₄ phosphor.^35,36 The corresponding results are summarized in Table 1.

(3) Yb³⁺–Er³⁺ energy transfer under infrared excitation: at high temperature, fluorescence quenching of emission bands of Er³⁺ ions induced by the temperature was obvious at 423 K.^21–35 It is necessary to enhance the emission intensity of phosphors in the process research of optical temperature sensing.

Among the rare-earth ions, the Yb³⁺ ion consists of only two levels, and has only one electronic excited state, ²F_5/2, that is located in the near infrared region at about 980 nm.¹³ Moreover, high power InGaAs diode lasers are available to directly pump the Yb³⁺ absorption band around 980 nm. Fluorescent materials with Yb³⁺ as a sensitizer can convert short infrared into visible/ultraviolet light via energy transfer between lanthanide ions. As reported, the doping content of Yb³⁺ was usually kept at 18% or higher, because the large energy gap between the excited state ²F_5/2 and ground state ²F_7/2 blocks multiphoton cross-relaxation.¹⁴ Thus, in order to enlarge the photo-absorption cross section of 980 nm infrared light, the Yb³⁺ ion is chosen to be a sensitizer for the Er³⁺ ion. Under 980 nm excitation, the two successive energy transfers from Yb³⁺ to Er³⁺ can induce the population of the thermally coupled ²H_11/2 and ⁴S_3/2 levels of Er³⁺,^37,39–50 and five successive energy transfers from Yb³⁺ to Er³⁺ can induce the population of the thermally coupled ⁴D_7/2 and ⁴G_9/2 levels of Er³⁺,³⁸ as shown in Fig. 3(c). The temperature dependent fluorescence intensity ratios, I_U/I_L and I^′_U/I^′_L, are used to evaluate the optical temperature sensing properties.

Recently, in the low temperature range, based on green luminescence emissions of Er³⁺–Yb³⁺ co-doped NaYF₄ nanoparticles, Vetrone et al. demonstrated the optical temperature sensing properties in the internal temperature of the living HeLa cervical cancer cell from 25 °C to 45 °C.³⁷ Notably, the temperature dependence of the five-photon 256 nm (⁴D_7/2 → ⁴I_15/2) and 276 nm (⁴G_9/2 → ⁴I_15/2) ultraviolet up-conversion luminescence in Yb³⁺–Er³⁺ co-doped β-NaLuF4 nanocrystals was studied firstly by Zheng et al. from 303 K to 523 K with a maximum sensitivity of 0.0052 K⁻¹ at 303 K,³⁸ as shown in Fig. 5.

Fig. 5 Temperature dependent (a) spectrum of Er³⁺ ions from ⁴D_7/2 and ⁴G_9/2 levels; (b) log plot of the FIR; (c) FIR relative to the temperature; and (d) sensor sensitivity under 980 nm excitation. Reprinted from ref. 38 with permission of Optical Society of America.

In the high temperature range, special phosphors are needed with high thermal stability and high solubility between hosts and lanthanide ions. Among the oxide materials, Y₂O₃ is a good candidate host, due to its wide bandgap, high melting point, high solubility between Y³⁺ and Er³⁺, and good transparency from ultraviolet to infrared. Recently, in Er³⁺–Yb³⁺ co-doped Y₂O₃ nanoparticles, optical temperature sensing properties were studied in the temperature range from 93 K to 613 K through analyzing the temperature dependent fluorescence intensity ratio of the two green emissions.^39–42 A maximum sensitivity of 528 × 10⁻⁴ K⁻¹ for 150 K was obtained in the Y_1.97Yb_0.02Er_0.01O₃ nanophosphors.⁴⁰ The fluorescence intensity ratio of the green up-conversion emissions at 523 nm and 545 nm in Er³⁺–Yb³⁺ co-doped Al₂O₃ was studied by Dong et al. as a function of temperature with a maximum sensitivity of 0.0051 K⁻¹ at 495 K.⁴³ Additionally, oxide salts, such as BaMoO₄, Y₂SiO₅, MWO₄ (M = Ca, Sr), YVO₄, BaTiO₃, Yb₂Ti₂O₇, Bi₇Ti₄NbO₂₁, and LiNbO₃, were doped with Er³⁺ and Yb³⁺ ions, and were studied as optical temperature sensors by the temperature dependent fluorescence intensity ratio of the two green emissions.^43–54 From Table 1, one can find that the maximum S_R, 0.0528 K⁻¹, is obtained at 150 K in Er³⁺–Yb³⁺ co-doped Y₂O₃ phosphors under 978 nm excitation, and the maximum S_A, 2109.31/T², is obtained in Er³⁺–Yb³⁺ co-doped CaWO₃ phosphors under 978 nm excitation.

The term δ is a key parameter used to judge the calculation accuracy of the fluorescence intensity ratio technique in the data fitting process. ΔE_f and ΔE_m between the TCL of Er³⁺ ions and the corresponding δ values are calculated and summarized in Table 2 through analyzing ref. 21–55. In Table 2, we can see that most of the Er³⁺ doped phosphors have a δ value of more than 5%, and the maximum δ value is about 53.89% for the Er³⁺ doped Si–B–Ba–Na glass. The δ values less than 5% are obtained in Er³⁺ doped tellurite glass, Na_0.82Ca_0.08Er_0.16Y_0.853F₄ bulks, PKAZLF glass, and ZBLALiP glass, respectively. The small δ values for tellurite, PKAZLF, and ZBLALiP glass materials can be explained as follows: the successive excited state absorption overcomes the energy transfer between the thermally coupled levels and other levels, due to the abnormal local ligand fields around the Er³⁺ sites in the tellurite, PKAZLF, and ZBLALiP glass hosts.^14,17 The population of the TCL of Er³⁺ ions at high temperature obeys the Boltzmann distribution. As shown in Table 2, most of the Er³⁺–Yb³⁺ co-doped phosphors have a value of δ of more than 10%, and the maximum δ value is about 90% for the Er³⁺–Yb³⁺ co-doped β-NaLuF₄ nanocrystals. Values of δ of less than 5% are observed in Er³⁺–Yb³⁺ co-doped Y₂O₃, and Er³⁺–Mo⁶⁺ co-doped Yb₂Ti₂O₇. The large δ value may be ascribed to the actual deviation of eqn (6) induced by energy transfer from the TCL to the other excited states.¹⁴

Table 2 Host dependent ΔE_f, ΔE_m, and δ in Er³⁺ doped and Yb³⁺–Er³⁺ co-doped phosphors

Samples	ΔEf (cm^−1)	ΔEm (cm^−1)	δ (%)	Ref.
Er^3+ doped tellurite glass	781	802	2.62	21
Er^3+ doped In–Zn–Sr–Ba glass	861	771.8	11.55	22
Er^3+ doped α-NaYF4 glass ceramics	741	865.8	14.41	23
Er^3+ doped PKAZLF glass	873	902.2	3.24	24
Er^3+ doped Te–Pb–Al glass	810	686	18.1	25
Er^3+ doped Sr–Ba–Nb–B glass	872.3	748	16.62	26
Er^3+ doped Te–Pb–Al glass	765	810	5.56	27
Er^3+ doped ZBLALiP glass	681	700	2.72	28
Er^3+ doped Ba(Zr,Ca)TiO3 ceramics	789	850	7.18	29
Er^3+ doped Si–B–Ba–Na glass	236	511.7	53.89	30
Er^3+ doped BaTiO3 nanocrystals	662.4	729.9	9.25	31
Er^3+ doped PLZT ceramics	773	1027.5	24.77	32
Er^3+, Mo^6+ co-doped YbAG	634.8	842.1	24.62	33
Er^3+ doped ZnO	611	573.5	6.54	34
Er^3+ doped chalcogenide glass	800	849.9	5.87	35
Na0.82Ca0.08Er0.16Y0.853F4	675.3	670.27	0.75	36
Er^3+, Yb^3+ co-doped α-NaYF4	724.4	766.1	5.44	37
Er^3+, Yb^3+ co-doped β-NaLuF4	270.6	2830.6	90.44	38
Er^3+, Yb^3+, Eu^3+ tri-doped Y2O3	738.54	800	7.68	39
Er^3+, Yb^3+ co-doped Y2O3	826.9	810.6	2.01	40
Er^3+, Yb^3+ co-doped Y2O3	615.7	800	23.0	41
Er^3+, Yb^3+, Eu^3+ tri-doped Y2O3	474.9	971	51.09	42
Er^3+, Yb^3+ co-doped Al2O3	679.4	771.8	11.97	43
Er^3+, Yb^3+ co-doped BaMoO4	607	716	15.22	44
Er^3+, Yb^3+ co-doped Y2SiO5	781	686.1	13.83	45
Er^3+, Yb^3+, Nd^3+ tri-doped Y2SiO5	498	601.7	17.23	46
Er^3+, Yb^3+ co-doped SrWO4	602	766.1	21.42	47
Er^3+, Yb^3+ co-doped YVO4	538	693	22.36	48
Er^3+, Yb^3+ co-doped Na0.5Bi0.5TiO3	706.8	800	11.65	49
Er^3+, Yb^3+ co-doped NaBiTiO3	574.8	800	28.15	50
Er^3+, Yb^3+ co-doped CaWO4	1455	1530	4.90	51
Er^3+, Mo^6+ co-doped Yb2Ti2O7	478.6	482	0.71	52
Er^3+, Yb^3+ co-doped Bi7Ti4NbO21	775	820.7	5.57	53
Er^3+, Yb^3+ co-doped LiNbO3	860	686.2	25.33	54
Er^3+, Yb^3+ co-doped TeO2–WO3 glass	678.94	826.5	17.85	55

---

3.2 Optical thermometry based on Yb³⁺–Tm³⁺ co-doped phosphors

The trivalent Tm³⁺ ion has a 4f¹² electronic configuration, and is reported as one of the most efficient blue luminescence center ions. It has two couples of adjacent thermally coupled levels, such as ³F_2,3 and ³H₄, and ¹G_4(a) and ¹G_4(b), whose relative emission intensity is strongly temperature dependent. As shown in Fig. 6, the energy transfer (ET) from Yb³⁺ to Tm³⁺ is the main population mechanism of optical thermometry based on up-conversion luminescence of the Tm³⁺ ion.⁶⁰ Under 980 nm infrared excitation, the thermally coupled levels ¹G_4(a) and ¹G_4(b) are populated by three successive ETs from Yb³⁺ to Tm³⁺. The thermally coupled levels ³F_2,3 and ³H₄ are populated by two successive ETs from Yb³⁺ to Tm³⁺. The ions in the high energy excited states, such as ¹G_4(a) and ¹G_4(b), ³F_2,3 and ³H₄, relax to the next lower energy levels through non-radiative relaxation, giving two blue emissions with the intensities I_U and I_L, and two infrared emissions with the intensities I^′_U and I^′_L. The temperature dependent fluorescence intensity ratios, I_U/I_L and I^′_U/I^′_L, are used to evaluate the optical temperature sensing properties.

Fig. 6 The mechanism of optical thermal sensing through Yb³⁺–Tm³⁺ energy transfer under 980 nm infrared excitation.

The spectral properties of the Tm³⁺ luminescence band at 480 nm from the ¹G₄ → ³H₆ transition in Tm³⁺ and Yb³⁺ co-doped NaNbO₃ nanocrystals were analyzed in the biophysical temperature range from 297 K to 353 K with a 980 nm excitation source.⁵⁶ Novel TCL, ¹G_4(a) and ¹G_4(b), were observed firstly by Pereira et al. Experiments proved that the population re-distribution among the thermally coupled stark levels could be successfully used for ratiometric thermal sensing. By analyzing the R value between the 700 nm and 800 nm up-conversion emissions, the optical temperature sensing properties of the Tm³⁺/Yb³⁺ co-doped oxyfluoride PbF₂ glass ceramic were studied by Xu et al. in the temperature range of 293–703 K.⁵⁷ The optical temperature sensing properties of Tm³⁺ were studied in Tm³⁺–Yb³⁺ co-doped Y₂O₃ bulk and Y₂O₃ sub-micronic spherical particles.^58,59 The S_A and S_B values obtained from the Tm³⁺–Yb³⁺ co-doped Y₂O₃ sub-micronic spherical particles are larger than those from the Tm³⁺–Yb³⁺ co-doped Y₂O₃ bulk.

Recently, trivalent rare-earth ion doped fluoride nanocrystals were reported as the most efficient up-conversion materials.^13,14 Based on ultraviolet up-conversion emissions from the ⁶P_5/2/⁸S_7/2 and ⁶P_7/2/⁸S_7/2 TCL, Zheng et al. demonstrated firstly the optical temperature sensing properties in Tm³⁺–Yb³⁺–Gd³⁺ tri-doped NaLuF₄ microcrystals in the range of 298–523 K,⁶⁰ as shown in Fig. 7. The maximum sensor sensitivity of about 0.0029 K⁻¹ was found at 298 K. The maximum sensor sensitivities were found to be about 0.0004 K⁻¹ (333 K) and 0.0029 K⁻¹ (298 K) by analyzing the ultraviolet emissions from the ⁶P_5/2/⁸S_7/2 and ⁶P_7/2/⁸S_7/2 levels, respectively. To our knowledge, fluoride nanocrystals are easily oxidized at high temperature. To overcome it, NaYbF₄:Tm³⁺@SiO₂ core–shell materials were synthesized, and their optical temperature sensing properties based on the ³F₂/³H₄ TCL of the Tm³⁺ ion were studied by Wang et al. in a wide temperature range from 100 K to 700 K.⁶¹ A better behavior as a low temperature sensor has been obtained with a minimum sensitivity of 5.4 × 10⁻⁴ K⁻¹ at 430 K.

Fig. 7 Temperature dependent (a) ultraviolet spectra in the range of 276–281 nm, and (b) R₁ of the emissions from the transitions of ⁶I_9/2/⁸S_7/2 and ⁶I_7/2/⁸S_7/2 of NaLuF4:Yb³⁺, Tm³⁺, Gd³⁺ microcrystals. Temperature dependent (c) ultraviolet spectra in the range of 304–318 nm, and (d) R₂ of the emissions from the transitions of ⁶P_5/2/⁸S_7/2 and ⁶P_7/2/⁸S_7/2 of NaLuF₄:Yb³⁺, Tm³⁺, Gd³⁺ microcrystals. Reproduced from ref. 60 with permission of the Royal Society of Chemistry.

In Table 3 we summarize, for the sake of comparison, the different performance parameters of sensitivities and the involved transitions of Tm³⁺ doped phosphors. The maximum value of S_R, 0.078 K⁻¹, is obtained in Tm³⁺–Yb³⁺ co-doped Y₂O₃ sub-micronic spherical particles. The maximum value of S_A, 2829.5/T², is obtained in Tm³⁺/Yb³⁺ co-doped oxyfluoride PbF₂ glass ceramic. To study the feasibility evaluation in application, ΔE_f, ΔE_m, and δ are studied in Tm³⁺–Yb³⁺ co-doped NaNbO₃ nanocrystals, oxyfluoride PbF₂ glass ceramic, Y₂O₃, NaLuF₄ microcrystals, and NaYbF₄ under 980 nm excitation, as shown in Fig. 8. A large δ value of more than 80% is observed in Tm³⁺–Yb³⁺ co-doped NaNbO₃ nanocrystals. The large δ value may be ascribed to the actual deviation of eqn (6) induced by energy transfer from the TCL to the other excited states.¹⁴ It is accurate to use NaYbF₄:Tm³⁺@SiO₂ core–shell materials and Tm³⁺/Yb³⁺ co-doped oxyfluoride Si–Pb glass ceramic to evaluate the scale of optical temperature sensing.

Table 3 The maximum sensitivity values of Tm³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Tm^3+, Yb^3+	NaNbO3	976	480, 486	^1G4, ^3F2,3, ^3H4 → ^3H6	293–353	0.0008 °C^−1 (25 °C)	93.53/T^2	56
Tm^3+, Yb^3+	PbF2 glass ceramics	980	700, 800	^3F2,3, ^3H4 → ^3H6	293–703	0.0006 K^−1 (1360 K)	2829.5/T^2	57
Tm^3+, Yb^3+	Y2O3	976	476, 488	^1G4(a), ^1G4(b) → ^3H6	303–753	0.0035 K^−1 (303 K)	452.51/T^2	58
Tm^3+, Yb^3+	Y2O3	978	454, 815	^1D2 → ^3F4, ^3H4 → ^3H6	10–300	0.078 K^−1 (270 K)	566.91/T^2	59
Tm^3+, Yb^3+, Gd^3+	NaLuF4	980	307, 312.4	^6P5/2, ^6P7/2 → ^8S7/2	298–523	0.0004 K^−1 (333 K)	667/T^2	60
		980	277.3, 279.9	^6I9/2, ^6I7/2 → ^8S7/2	298–523	0.0029 K^−1 (298 K)	408/T^−2	60
Tm^3+	NaYbF4@SiO2	980	697, 798	^3F2,3, ^3H4 → ^3H6	100–700	0.00054 K^−1 (100 K)	2677.39/T^2	61

---

Fig. 8 Host dependent ΔE_f and ΔE_m between the TCL of Tm³⁺ ions, and the corresponding δ values.

3.3 Optical thermometry based on Ho³⁺ doped phosphors

The trivalent Ho³⁺ ion has a 4f¹⁰ electronic configuration, and is reported as one of the most efficient green luminescence center ions, due to the intermediate ⁵S₂ level with a long fluorescence lifetime.^13,14 It has three couples of adjacent TCL, such as ⁵F₄/⁵S₂, ⁵F_2,3/³K₈, and ⁵G₆/⁵F₁. As shown in Fig. 9, the two-step up-conversion of Ho³⁺ and successive energy transfer from Yb³⁺ to Ho³⁺ are the main mechanisms of optical thermometry based on Ho³⁺ doped phosphors.^60–68 In the case of a low doping concentration (<1%), the ESA process refers to a sequential absorption of two pump infrared photons at 890 nm by a single Ho³⁺ ion. Upon 980 nm laser irradiation, Yb³⁺ absorbs infrared photons with the generation of ²F_7/2 → ²F_5/2 upward transitions. Subsequently, it donates the energy to the adjacent Ho³⁺ through a phonon assisted energy transfer, with Yb³⁺ dropping back to its ²F_7/2 ground state. This promotes the Ho³⁺ ions to their excited states, such as ⁵I₆, ⁵S₂, and ⁵G₅. The thermally coupled ⁵F₄/⁵S₂, ⁵F_2,3/³K₈, and ⁵G₆/⁵F₁ energy levels are populated by the process of non-radiative relaxation. The fluorescence intensity ratios, I_U/I_L, I^′_U/I^′_L, and I^′′_U/I^′′_L, are dependent strongly on temperature, and are used to study the optical temperature sensing properties.

Fig. 9 The mechanism of optical thermal sensing through (a) up-conversion of the Ho³⁺ ion under 890 nm infrared excitation, and (b) Yb³⁺–Ho³⁺ energy transfer under 980 nm infrared excitation.

It was reported that there were two formulas used to fit ΔE_f of the Ho³⁺ ions, due to the two different transitions, such as from the TCL to the other excited state, and from the TCL to the ground state. One formula is eqn (6), which is suitable to fit ΔE_f corresponding to transitions from ⁵F₄/⁵S₂, ⁵F_2,3/³K₈, and ⁵G₆/⁵F₁ to the ⁵I₈ ground state. Using eqn (6) as a fitting formula, the temperature dependent fluorescence intensity ratios were studied in Ho³⁺ doped TeO₂ glass and LiTeO₂ glass under 890 nm excitation,^62,63 and Ho³⁺–Yb³⁺ co-doped CaMoO_4, CaWO₄, and PbF₂ glass ceramic under 980 nm excitation.^66–68 A maximum S_R of 0.098 K⁻¹ for 130 K was obtained in Ho³⁺ doped TeO₂ glass, and a maximum S_A of 2191.9/T² was obtained in Ho³⁺–Yb³⁺ co-doped glass ceramic.

The other formula is as follows:

where C₁, C₂ and C₃ are constants that depend on spontaneous emission rates, degeneracy of the energy levels and emission energies.⁶⁰ Eqn (11) is used to fit ΔE_f originating from the transitions of ⁵F₄ → ⁵I₈ and ⁵S₂ → ⁵I₇. Using eqn (11) as a fitting formula, the temperature dependent fluorescence intensity ratios were studied in Ho³⁺ doped In–Zn–Sr–Ba glass under 473 nm excitation,²² Ho³⁺–Yb³⁺ co-doped Y₂O₃ under 978 nm excitation,⁶⁰ and Ba_0.77Ca_0.23TiO₃ and Ca₁₂Al₁₄O₃₃ under 980 nm excitation.^64,65 The Ho³⁺–Yb³⁺ co-doped Y₂O₃ powder was obtained via the spray pyrolysis method; its SEM image in Fig. 10(a) shows that the resulting particles were spherical, sub-micronic in size and un-agglomerated. The up-conversion emission spectra in Fig. 10(b) show that the green (536 nm) and infrared (758, 764, and 772 nm) emissions were dependent strongly on temperature. Fig. 10(c) shows the experimental fluorescence intensity ratio (FIR) of emission at 536 nm relative to the emissions at 758 nm, 764 nm and 772 nm, and these were fitted with eqn (11). At temperatures of 85, 84 and 90 K, the sensitivity of Ho³⁺ in Fig. 10(d) exhibits maximum values of 0.097, 0.065 and 0.046 K⁻¹ for emissions at 536/772, 536/764 and 536/758 nm, respectively. The values of fitted ΔE in the inset of Fig. 10(c) and sensitivity are dependent strongly on the FIR.

Fig. 10 (a) SEM image, (b) up-conversion spectra under 978 nm excitation, (c) FIR of emission at 536 nm relative to 758 nm, 764 nm and 772 nm emissions, and (d) the temperature dependence of sensitivity of Y₂O₃:Yb³⁺, Ho³⁺ particles. Reproduced from ref. 59 with permission of Elsevier.

In Table 4 we summarize, for the sake of comparison, the different performance parameters of Ho³⁺ doped and Ho³⁺–Yb³⁺ co-doped phosphors. The maximum value of S_R, 0.0098 K⁻¹, is obtained in Ho³⁺–Yb³⁺ co-doped TeO₂ glass. The maximum value of S_A, 2191.9/T², is obtained in the Tm³⁺/Yb³⁺ co-doped oxyfluoride PbF₂ glass ceramic.

Table 4 The maximum sensitivity values of Ho³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Ho^3+	In–Zn–Sr–Ba glass	473	545, 750	^5F4/^5S2 → ^5I8, ^5I7	20–300	0.0036 K^−1 (59 K)	181.64/T^2	22
Ho^3+	TeO2 glass	890	538, 545	^5F4, ^5S2 → ^5I8	265–440	0.0098 K^−1 (130 K)	255/T^2	62
Ho^3+	LiTeO2 glass	890	538, 543	^5F4, ^5S2 → ^5I8	265–383	0.0063 K^−1 (265 K)	255/T^2	63
Ho^3+, Yb^3+	Y2O3	978	536.5, 772.6	^5F4/^5S2 → ^5I8, ^5I7	10–300	0.0097 K^−1 (85 K)	241.2/T^2	60
Ho^3+, Yb^3+	Ba0.77Ca0.23TiO3	980	546, 754	^5F4/^5S2 → ^5I8, ^5I7	93–300	0.0053 K^−1 (93 K)	182.3/T^2	64
Ho^3+, Yb^3+	Ca12Al14O33	980	467, 492, 542, 552	^5G4/^5G5, ^5F4/^5S2 → ^5I8	298–500			65
Ho^3+, Yb^3+	CaMoO4	980	460, 489	^5F3, ^3K8 → ^5I8	303–543	0.0066 K^−1 (353 K)	648.5/T^2	66
Ho^3+, Yb^3+	CaWO4	980	460, 487	^5G6/^5F1, ^5F2,3/^3K8 → ^5I8	303–923	0.0050 K^−1 (923 K)	1890/T^2	67
Ho^3+, Yb^3+	PbF2 glass ceramic	980	445, 485	^5G6/^5F1, ^5F2,3/^3K8 → ^5I8	303–643	0.00102 K^−1 (1119 K)	2191.9/T^2	68

---

To value the fitting accuracy, ΔE_f and ΔE_m between the TCL of the Ho³⁺ ions and the corresponding δ values are calculated in Fig. 11 through analyzing ref. 22, 60 and 62–68. The thermally coupled ⁵F₄/⁵S₂ energy levels were studied as a temperature function in ref. 22, 60 and 62–65. The minimum δ value of 5.09% is obtained in Ho³⁺ doped LiTeO₂ glass. The thermally coupled ⁵F_2,3/³K₈ energy levels were studied as a temperature function in ref. 66. A large δ value of 64.6% was obtained in Ho³⁺–Yb³⁺ co-doped CaMoO₄. The thermally coupled ⁵G₆/⁵F₁ energy levels were studied in ref. 67 and 68. A small δ value of 8.2% was obtained in Ho³⁺–Yb³⁺ co-doped CaWO₄. The large δ value may be ascribed to the lack of correction on eqn (11). If eqn (11) is used directly to fit ΔE_f, energy transfers from the TCL to the other excited states will be neglected. In fact, when the Ho³⁺ concentration is high enough, the cross-relaxation process, ⁵S₂ + ⁵I₈ → ⁵I₄ + ⁵I₇, occurs frequently among Ho³⁺ ions.¹³

Fig. 11 Host dependent ΔE_f and ΔE_m between the TCL of Ho³⁺ ions, and the corresponding δ values.

3.4 Optical thermometry based on Nd³⁺ doped phosphors

Among the rare-earth ions, Nd³⁺ has a 4f³ electronic configuration, and is reported as one of the most efficient activated ions to obtain laser emissions. It is significant work to study the optical thermal sensing of Nd³⁺ doped phosphors, since the temperature of the laser crystals can be monitored through analyzing the temperature dependent spectrum of Nd³⁺. Nd³⁺ has three couples of adjacent thermally coupled levels, ⁴F_5/2/⁴F_3/2 (TCL1), ⁴F_7/2/⁴F_3/2 (TCL2), and ⁴F_7/2/⁴F_5/2 (TCL3). As shown in Fig. 12, the mechanism of optical thermometry of Nd³⁺ doped phosphors is up-conversion and energy transfer from Yb³⁺ to Nd³⁺ assisted with phonons.^69–71 It needs one 532 nm green photon to populate the thermally coupled ⁴F_5/2 and ⁴F_3/2 energy levels through non-radiative relaxation. Upon 980 nm laser irradiation, Yb³⁺ absorbs infrared photons with the generation of ²F_7/2 → ²F_5/2 upward transitions. Subsequently, it donates the energy to the adjacent Nd³⁺ through a phonon assisted energy transfer, with Yb³⁺ dropping back to its ²F_7/2 ground state. This promotes the Nd³⁺ ions to their excited state, ⁴F_3/2. The ⁴F_5/2 energy level is populated by thermal excitation. The fluorescence intensity ratio, I_U/I_L, is dependent strongly on temperature, and is used to study the optical temperature sensing properties.

Fig. 12 The mechanism of optical thermal sensing through (1) down-conversion of the Nd³⁺ ion under 532 nm green excitation, and (2) Yb³⁺–Nd³⁺ energy transfer under 980 nm infrared excitation.

Recently, based on down-conversion luminescence emissions centered at 810 nm and 880 nm of Nd³⁺ ions, optical temperature sensing in the temperature range from 300 K to 850 K was studied by Rodríguez et al. in Nd³⁺ doped phosphate glass under 532 nm excitation.⁶⁹ Using a 980 nm diode laser as an excitation source, the temperature sensing properties based on the infrared luminescence emissions at 755 nm, 805 nm, and 872 nm of Nd³⁺/Yb³⁺ co-doped CaWO₄ powders has been discussed by Xu et al. in the temperature range from 303 K to 873 K,⁷⁰ as shown in Fig. 13(a). Three TCL, such as TCL1, TCL2, and TCL3, were proven by using the fluorescence intensity ratio technique, as shown in Fig. 13(b). Similarly, the temperature sensing properties based on near infrared emissions from TCL1, TCL2, and TCL3 of Nd³⁺/Yb³⁺ co-doped oxyfluoride glass ceramic containing PbF₂ nanocrystals were studied by Xu et al. as a function of temperature in the range of 303–623 K,⁷¹ as shown in Fig. 13(c) and (d). In Table 5 we summarize, for the sake of comparison, the different performance parameters of Nd³⁺ doped and Nd³⁺–Yb³⁺ co-doped phosphors.

Fig. 13 Temperature dependent (a) near infrared emission spectra and (b) fluorescence intensity ratio (FIR) in Nd³⁺/Yb³⁺ co-doped CaWO₄ phosphors excited by a 980 nm laser. Temperature dependent (c) near infrared emission spectra and (d) FIR in Nd³⁺/Yb³⁺ co-doped oxyfluoride glass ceramic containing PbF₂ nanocrystals excited by a 980 nm laser. Reproduced from ref. 70 and 71 with permission of Optical Society of America and Elsevier.

Table 5 The maximum sensitivity values of Nd³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Nd^3+	P–K–Ba–Al glass	532	810, 880	^4F5/2, ^4F3/2 → ^4I9/2	300–850	0.0153 K^−1 (300 K)	1306.47/T^2	69
Nd^3+, Yb^3+	CaWO4	980	755, 872	^4F7/2, ^4F3/2 → ^4I9/2	303–873		2746.5/T^2	70
			805, 872	^4F5/2, ^4F3/2 → ^4I9/2			1458.5/T^2
			755, 805	^4F7/2, ^4F5/2 → ^4I9/2			1336.7/T^2
Nd^3+, Yb^3+	PbF2 glass	980	750, 863	^4F7/2/^4S3/2, ^4F3/2 → ^4I9/2	303–623		3010.07/T^2	71
	Ceramic		750, 803	^4F7/2/^4S3/2, ^2H9/2/^4F5/2 → ^4I9/2			1884.66/T^2
			803, 863	F5/2/^2H9/2, ^4F3/2 → ^4I9/2			1763.23/T^2

---

The ΔE_f and ΔE_m values between the TCL of Nd³⁺ ions and the corresponding δ values are calculated in Fig. 14 through analyzing ref. 69–71. The temperature dependent emissions at 810 nm and 880 nm from TCL1 were studied in Nd³⁺ doped phosphate glass (ref. 69), and a small δ value of 2.55% was obtained. The temperature dependent emissions at 755 nm, 805 nm, and 872 nm from TCL1, TCL2, and TCL3 were studied in the Nd³⁺–Yb³⁺ co-doped CaWO₄ powders (ref. 70). The δ values corresponding to TCL1, TCL2 and TCL3 are calculated to be 6.58%, 5.34%, and 12.03%. The temperature dependent emissions at 750 nm, 803 nm, and 863 nm from TCL1, TCL2, and TCL3 were studied in the Nd³⁺–Yb³⁺ co-doped oxyfluoride glass ceramic (ref. 71). The corresponding δ values are calculated to be 8.69%, 27.45%, and 26.67%. If TCL1 and TCL3 are used to study the optical temperature sensing properties, we can get small and large δ values, respectively. It may be ascribed to the actual deviation of eqn (6) induced by the cross-relaxation process between the Nd³⁺ ions, ⁴F_7/2 + ⁴I_9/2 → ⁴F_3/2 + ⁴I_11/2.¹³

Fig. 14 Host dependent ΔE_f and ΔE_m between the TCL of Nd³⁺ ions, and the corresponding δ values.

3.5 Optical thermometry based on Dy³⁺ doped phosphors

The Dy³⁺ ion with a 4f⁹ electronic configuration has a couple of adjacent TCL, ⁴I_5/2 and ⁴F_9/2. As shown in Fig. 15, the mechanism of optical thermometry based on luminescence of Dy³⁺ ions is down-conversion induced by ultraviolet excitation.^72,73 Under ultraviolet excitation, Dy³⁺ ions are excited directly to the higher excited states by the GSA transition. The ions in the excited states relax to the next lower energy level, ⁴I_5/2 and ⁴F_9/2, through non-radiative relaxation. Three emissions with the intensities I_U, I_L, and I^′_L occur by the radiative transitions from ⁴I_5/2/⁴F_9/2 to ⁶H_13/2 and ⁶H_15/2. The luminescence intensity ratios, I_U/I_L and I_U/I^′_L, change with temperature, and are the precise evaluation scale of optical temperature sensing.

Fig. 15 The mechanism of optical thermal sensing through down-conversion of the Dy³⁺ ion under 355 nm ultraviolet excitation.

Optical temperature sensing of the Dy³⁺ doped BaYF₅ nanoparticles was studied by Cao et al. in the temperature range from 293 K to 773 K.⁷² The temperature dependent fluorescence spectra of Dy³⁺ ions under excitation of a 355 nm laser in Fig. 16(a) show that the intensity of the 455 nm emission increased with a temperature increase, and the intensity of the 478 nm emission was independent of temperature. The temperature dependent luminescence intensity ratio of the 455 nm and 478 nm emissions was fitted in Fig. 16(b), in which the line fitted very well with the experimental data at a temperature higher than 380 K. By analyzing the temperature dependent 455 nm and 481 nm blue emissions, Boruc et al. demonstrated optical temperature sensing of Dy³⁺ doped Y₄Al₂O₉ crystals.⁷³ Temperature dependent emission spectra in Fig. 16(c) show that the intensity of the 455 nm emission increased with a temperature increase, and the intensity of the 481 nm emission decreased with a temperature increase. Temperature dependent luminescence intensity ratios of two blue emissions in Fig. 16(d) show different sensitivity values in two temperature ranges, such as 296 K < T < 573 K and 573 K < T < 973 K. The performance parameters of Dy³⁺ doped BaYF₅ and Y₄Al₂O₉ are summarized in Table 6. To value the fitting accuracy, ΔE_f and ΔE_m between the TCL of Dy³⁺ ions and the corresponding δ values are calculated through analyzing ref. 72 and 73. The δ value for the Dy³⁺ doped BaYF₅ nanoparticles is 1.09%. For the Dy³⁺ doped Y₄Al₂O₉ crystals, the δ value is 18.8% at the temperature range 296 K < T < 573 K, and it is 11.8% at the temperature range 573 K < T < 973 K. It is accurate to use the Dy³⁺ doped BaYF₅ nanoparticles to evaluate the scale of optical temperature sensing.

Fig. 16 Temperature dependent (a) emission spectra and (b) fluorescence intensity ratio (FIR) between the 455 nm and 478 nm emissions for the BaYF₅:2%Dy³⁺ sample. Temperature dependent (c) emission spectra and (d) Arrhenius plot of the intensity ratio for Dy³⁺ doped Y₄Al₂O₉ crystals. Reproduced from ref. 72 and 73 with permission of Optical Society of America and Elsevier.

Table 6 The maximum sensitivity values of Dy³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Dy^3+	BaYF5	355	455, 478	^4I5/2, ^4F9/2 → ^6H15/2	293–773	0.02 K^−1 (700 K)	1507.07/T^2	72
Dy^3+	Y4Al2O9	355	455, 481	^4I5/2, ^4F9/2 → ^6H15/2	296–573		1438.7/T^2	73
					573–973	0.003 °C^−1(973 K)	1937.6/T^2

---

3.6 Optical thermometry based on Eu³⁺ doped phosphors

Among the rare-earth ions, Eu³⁺ has a 4f⁶ electronic configuration, and is reported as one of the most efficient activated ions to realize red emission in light emitting diodes (LEDs).^74–76 It is very significant to study optical thermal sensing of Eu³⁺ doped phosphors, since the temperature of LEDs can be monitored through analyzing the temperature dependent spectrum of Eu³⁺. Eu³⁺ has a couple of adjacent thermally coupled levels, ⁵D₁ and ⁵D₀. As shown in Fig. 17, the mechanism of optical thermometry based on luminescence of the Eu³⁺ ion is down-conversion induced by ultraviolet excitation.^77–80 Under ultraviolet excitation, Eu³⁺ ions are excited directly to the higher excited states by the GSA transition, and then relax to the next lower energy levels, ⁵D₁ and ⁵D₀, through non-radiative relaxation. Two emissions with the intensity of I_U and I_L occur by the radiative transitions from ⁵D₁/⁵D₀ to ⁷F_J (J = 0, 1, 2, 3, 4, 5, 6). At high temperature, for high Eu³⁺ concentration doped phosphors, the increase in the ⁵D₁ emissions may be attributed to a thermally assisted cross-relaxation (CR) process with the involvement of two Eu³⁺ ions initially in the excited ⁵D₀ and the ⁷F₃ state.⁷⁷

Fig. 17 The mechanism of optical thermal sensing through down-conversion of the Eu³⁺ ion under infrared excitation.

Meert et al. reported that the luminescence intensity ratios of the ⁵D₁ to ⁵D₀ emissions of the Eu³⁺ doped CaEu₂(WO₄)₄ scheelites changed in an exponential form with a temperature increase,⁷⁷ as shown in Fig. 18(a). The temperature dependent luminescence of spherical NaEuF₄ phosphors with different particle sizes was studied by Tian et al.⁷⁸ Fig. 18(b) shows the temperature dependent emission spectra of the NaEuF₄ spheres of 300 nm under 394 nm excitation. ΔE_f was calculated by fitting the temperature dependent emission intensity ratios of the ⁵D₁/⁵D₀ emissions. The sensitivity values of the spherical NaEuF₄ phosphors decreased with the size increase from 100 nm to 700 nm, and a sensitivity up to 0.43% K⁻¹ was achieved when the particle size was 100 nm. The temperature dependent luminescence spectrum of Sr₂CeO₄:Eu³⁺ was studied by Shi et al., as shown in Fig. 18(c). The integrated luminescence total intensity of the ⁵D₀ emissions at 591 nm, 616 nm, and 654 nm showed a line change relation in the temperature range from 373 K to 573 K.⁷⁹ Luminescence temperature sensing was studied by Nikolić et al. in Eu³⁺ doped TiO₂ nanoparticles over a temperature range of 307–533 K.⁸⁰ The 438 nm blue emission associated with the trap emission of the TiO₂ host and the emission peak of the Eu³⁺ ions at 613 nm are observed in Fig. 18(d) under continuous excitation at a wavelength of 360 nm. The temperature dependent luminescence intensity ratio of the 438 nm and 613 nm emissions was suitable to be used for temperature sensing. In Table 7 we summarize, for the sake of comparison, the different performance parameters of Eu³⁺ doped phosphors.

Fig. 18 (a) Emission spectrum of CaGd_1.8Eu_0.2(WO₄)₄ upon excitation at 395 nm. The inset is the integrated intensity ratio of the ⁵D₁ to ⁵D₀ emissions. (b) Temperature dependent emission spectra of NaEuF₄ spheres under 394 nm excitation. The inset is the emission intensity ratio of the ⁵D₁ to ⁵D₀ emissions with temperature. (c) Temperature dependent emission spectra of Sr₂CeO₄:Eu³⁺ under 365 nm excitation. The inset shows the integrated emission total intensity of 591 nm, 616 nm, and 654 nm at different temperatures. (d) Temperature dependent emission spectra of TiO₂:Eu³⁺ nanopowder. The inset is the emission intensity ratio of the 438 nm and 613 nm emissions with temperature. Reproduced from ref. 74–77 with permission of Optical Society of America and Elsevier.

Table 7 The maximum sensitivity values of Eu³⁺ doped phosphors by the fluorescence intensity ratio technique are presented, and the involved transitions from the TCL, excition wavelength, emission wavelength, and temperature range are included

Rare-earth ions	Host	λex (nm)	λem (nm)	Transitions	Temperature range (K)	SR (maximum)	SA	Ref.
Eu^3+	CaEu2(WO4)4 scheelite	395	535, 590	^5D1, ^5D0 → ^7F1	300–500	0.014 K^−1 (300 K)	1515.5/T^2	77
Eu^3+	NaEuF4 phosphor	394	500–560, 560–720	^5D1, ^5D0 → ^7F1	298–523	0.0043 K^−1	2398.2/T^2	78
Eu^3+	Sr2CeO4	365	591, 616, 654	^5D0 → ^7F1/2/3	303–573	1290 K^−1		79
Eu^3+	TiO2 nanoparticles	360	438, 613	Host trap, ^5D0 → ^7F2	307–533	0.0243 K^−1 (533 K)		80

---

To value the fitting accuracy, ΔE_f and ΔE_m between the TCL of the Eu³⁺ ions and the corresponding δ values are calculated through analyzing ref. 77 and 78. The δ value for the Eu³⁺ doped CaEu₂(WO₄)₄ scheelites is 38.7%, and the δ value for the spherical NaEuF₄ phosphors is 2.5%. As for the Eu³⁺ doped TiO₂ and Sr₂CeO₄, the values of δ are difficult to give, due to the incomplete data in ref. 76 and 77.

4. Conclusion and challenges

In summary, we have presented a detailed review of the Er³⁺, Tm³⁺, Ho³⁺, Nd³⁺, Dy³⁺, and Eu³⁺ doped phosphors to date for the achievement of optical temperature sensing from the analysis of temperature dependent spectra. It has been shown that many phosphors with small δ values can be used as basic light emitting materials for optical thermometry. In the case of Er³⁺, Tm³⁺, Ho³⁺, Nd³⁺, and Dy³⁺ doped phosphors (ΔE_m ≤ 900 cm⁻¹) thermal equilibrium can be assumed because the population of the thermally coupled levels follows a Boltzmann distribution. Beyond this limit changes in the emission intensities are due to the effect of “coupling/decoupling” at higher or lower temperatures. As in the case of the Eu³⁺ doped phosphors, the ⁵D₁ and ⁵D₀ levels (ΔE_f ∼ 1725 cm⁻¹) are thermally decoupled at lower temperatures, with thermalization at higher temperatures. With the significant fundamental scientific and technological knowledge already accumulated by past research, optical thermometry represents an encouraging prospect in the development of new temperature sensors. However, from the point of view of application, many challenges remain in the areas of materials science and physical mechanisms.

One related challenge is the issue of trying to find novel phosphor materials that can be used as temperature sensors at temperatures greater than 500 °C (773 K). Luminescence of rare-earth ion doped phosphors mentioned in this review is very weak at temperatures greater than 773 K, due to intense thermal quenching. It is difficult to measure temperature change in the range of more than 500 °C through the conventional optical temperature sensing technology. New phosphors with low thermal quenching rates and high luminescence efficiency should be synthesized cheaply. Instead of rare-earth ions, new luminescence centers with high luminescence intensity at high temperature (>773 K) may be excellent candidate activated ions in novel fluorescence sensors.

Another main challenge facing the field of optical thermometry is the lack of understanding of the physical mechanisms determining the transitions between the electronic levels of both the host lattice and the rare-earth ions. A theoretical model describing thermal population between the TCL in eqn (6) is suitable to be used in the data fitting process only in the cases without energy transfer. In fact, energy transfer between the host and the rare-earth ions becomes active at high temperature, and energy transfer among the rare-earth ions becomes active at a high doping concentration. Thus, the error values of ΔE_f for some phosphors mentioned in this review are large at more than 10%. In order to improve the calculation accuracy, a theoretical model describing the thermal population between the TCL in eqn (6) should be corrected for different phosphors. Additionally, the thermal affection of the excitation source should be considered in eqn (6).

Finally, we also believe that phosphor thermometry will be widely used in the future, not only for temperature detection in special environments, but also for some medical therapeutic equipment. In this sense, the development of phosphors with high thermal sensitivity and low thermal quenching of fluorescence will yield significant advances in the design of new optical temperature sensors. Colloidal phosphors with a random flightchain with biological tissues will be the target of innovation for disease diagnoses through the technology of optical temperature sensing.

Acknowledgements

This work was supported by National Natural Science Foundation of China (11404171, 11374162), Natural Science Youth Foundation of Jiangsu Province (BK20130865), the Six Categories of Summit Talents of Jiangsu Province of China (2014-XCL-021), and the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (14KJB430020).

References

1. F. Vetrone, R. Naccache, A. Zamarrón, A. J. de la Fuente, F. S. Rodríguez, L. M. Maestro, E. M. Rodriguez, D. Jaque, J. Solé and J. A. Capobianco, ACS Nano, 2010, 4, 3254–3258.
2. C. Gunawan, M. Lim, C. P. Marquis and R. Amal, J. Mater. Chem. B, 2014, 2, 2060–2083.
3. W. J. Liu and B. Z. Yang, Sens. Rev., 2007, 27, 298–309.
4. D. Jaque and F. Vetrone, Nanoscale, 2012, 4, 4301–4326.
5. C. D. S. Brites, P. P. Lima, N. J. O. Silva, A. M. An, V. S. Amaral, F. Palacio and L. D. Carlos, Nanoscale, 2012, 4, 4799–4829.
6. N. Armaroli and V. Balzani, Energy Environ. Sci., 2011, 4, 3193–3222.
7. E. D. Larson, G. Fiorese, G. J. Liu, R. H. Williams, T. G. Kreutz and S. Consonni, Energy Environ. Sci., 2010, 3, 28–42.
8. M. E. Borges, M. C. A. Galván, P. Esparza, E. Medina, P. M. Zarza and J. L. G. Fierro, Energy Environ. Sci., 2008, 1, 364–369.
9. L. Guo, W. P. Dong and S. T. Zhang, RSC Adv., 2014, 4, 41956–41967.
10. L. M. Maestro, C. Jacinto, U. R. Silva, F. Vetrone, J. A. Capobianco, D. Jaque and J. G. Solé, Small, 2011, 7, 1774–1778.
11. J. R. Lakowicz, I. Gryczynski, V. Bogdanov and J. Kusba, J. Phys. Chem., 1994, 98, 334–342.
12. P. Löw, B. Kim, N. Takama and C. Bergaud, Small, 2008, 4, 908–914.
13. H. Dong, L.-D. Sun and C.-H. Yan, Chem. Soc. Rev., 2015, 44, 1608–1634.
14. F. Wang and X. G. Liu, Chem. Soc. Rev., 2009, 38, 976–989.
15. S. G. Xiao, X. L. Yang, Z. W. Liu and X. H. Yan, J. Appl. Phys., 2004, 96, 1360.
16. M. D. Shinn, W. A. Sibley, M. G. Drexhage and R. N. Brown, Phys. Rev. B: Condens. Matter Mater. Phys., 1983, 27, 6635–6648.
17. S. A. Wade, S. F. Collins and G. W. Baxter, J. Appl. Phys., 2003, 94, 4743–4756.
18. M. Quintanilla, E. Cantelar, F. Cusso, M. Villegas and A. C. Caballero, Appl. Phys. Express, 2001, 4, 022601.
19. V. K. Rai, Appl. Phys. B: Lasers Opt., 2007, 88, 297–303.
20. S. F. L. Luis, U. R. R. Mendoza, E. Lalla and V. Lavín, Sens. Actuators, B, 2011, 158, 208–213.
21. G. Z. Sui, X. P. Li, L. H. Cheng, J. S. Zhang, J. S. Sun, H. Y. Zhong, Y. Tian, S. B. Fu and B. J. Chen, Appl. Phys. B: Lasers Opt., 2013, 110, 471–476.
22. P. H. González, S. F. L. Luis, S. G. Pérez and I. R. Martín, Mater. Res. Bull., 2011, 46, 1051–1054.
23. S. F. L. Luis, U. R. R. Mendoza, P. H. González, I. R. Martín and V. Lavín, Sens. Actuators, B, 2012, 174, 176–186.
24. N. Vijaya, P. Babu, V. Venkatramu, C. K. Jayasankar, S. F. L. Luis, U. R. R. Mendoza, I. R. Martín and V. Lavín, Sens. Actuators, B, 2013, 186, 156–164.
25. S. F. L. Luis, U. R. R. Mendoza, I. R. Martín, E. Lalla and V. Lavín, Sens. Actuators, B, 2013, 176, 1167–1175.
26. P. H. González, I. R. Martín, L. L. Martín, S. F. L. Luis, C. P. Rodríguez and V. Lavín, Opt. Mater., 2011, 33, 742–745.
27. S. F. L. Luis, U. R. R. Mendoza, E. Lalla and V. Lavín, Sens. Actuators, B, 2011, 158, 208–213.
28. Z. P. Cai and H. Y. Xu, Sens. Actuators, A, 2003, 108, 187–192.
29. P. Du, L. H. Luo, W. P. Li, Q. Y. Yue and H. B. Chen, Appl. Phys. Lett., 2014, 104, 152902.
30. C. R. Li, B. Dong, C. G. Ming and M. K. Lei, Sensors, 2007, 7, 2652–2659.
31. M. A. R. C. Alencar, G. S. Maciel and C. B. de Araújo, Appl. Phys. Lett., 2004, 84, 4753–4755.
32. A. S. S. de Camargo, J. F. Possatto, L. A. O. Nunes, E. R. Botero, E. R. M. Andreeta, D. Garcia and J. A. Eiras, Solid State Commun., 2006, 137, 1–5.
33. B. Dong, B. S. Cao, Y. Y. He, Z. Liu, Z. P. Li and Z. Q. Feng, Adv. Mater., 2012, 24, 1987–1993.
34. X. Wang, X. G. Kong, Y. Yu, Y. J. Sun and H. Zhang, J. Phys. Chem. C, 2007, 111, 15119–15124.
35. P. V. dos Santos, M. T. de Araujo, A. S. Gouveia-Neto, J. A. M. Neto and A. S. B. Sombra, IEEE J. Quantum Electron., 1999, 35, 395–399.
36. X. F. Wang, C.-S. Liu and X. H. Yan, RSC Adv., 2014, 4, 24170–24175.
37. F. Vetrone, R. Naccache, A. Zamarrón, A. J. de la Fuente, F. S. Rodríguez, L. M. Maestro, E. M. Rodriguez, D. Jaque, J. G. Solé and J. A. Capobianco, ACS Nano, 2010, 4, 3254–3258.
38. K. Z. Zheng, W. Y. Song, G. H. He, Z. Yuan and W. P. Qin, Opt. Express, 2015, 23, 7653–7658.
39. V. K. Rai, A. Pandey and R. Dey, J. Appl. Phys., 2013, 113, 083104.
40. V. Lojpur, G. Nikoli and M. D. Dramianin, J. Appl. Phys., 2014, 115, 203106.
41. P. Du, L. H. Luo, Q. Y. Yue and W. P. Li, Mater. Lett., 2015, 143, 209–211.
42. R. Dey, A. Pandey and V. K. Rai, Sens. Actuators, B, 2014, 190, 512–515.
43. B. Dong, D. P. Liu, X. J. Wang, T. Yang, S. M. Miao and C. R. Li, Appl. Phys. Lett., 2007, 90, 181117.
44. A. K. Soni, A. Kumari and V. K. Rai, Sens. Actuators, B, 2015, 216, 64–71.
45. N. Rakov and G. S. Maciel, Sens. Actuators, B, 2012, 164, 96–100.
46. N. Rakov and G. S. Maciel, Opt. Lett., 2014, 39, 3767–3769.
47. A. Pandey, V. K. Rai, V. Kumar, V. Kumar and H. C. Swart, Sens. Actuators, B, 2015, 209, 352–358.
48. M. K. Mahata, K. Kumar and V. K. Rai, Sens. Actuators, B, 2015, 209, 775–780.
49. P. Du and J. S. Yu, Ceram. Int., 2015, 41, 6710–6714.
50. P. Du, L. H. Luo, W. P. Li and Q. Y. Yue, J. Appl. Phys., 2014, 116, 014102.
51. W. Xu, Z. G. Zhang and W. W. Cao, Opt. Lett., 2012, 37, 4865–4867.
52. B. S. Cao, Y. Y. He, Z. Q. Feng, Y. S. Li and B. Dong, Sens. Actuators, B, 2011, 159, 8–11.
53. H. Zou, J. Li, X. S. Wang, D. F. Peng, Y. X. Li and X. Yao, Opt. Mater. Express, 2014, 4, 1545–1554.
54. M. Quintanilla, E. Cantelar, F. Cussó, M. Villegas and A. C. Caballero, Appl. Phys. Express, 2011, 4, 022601.
55. A. Pandey, S. Som, V. Kumar, V. Kumar, K. Kumar, V. K. Rai and H. C. Swart, Sens. Actuators, B, 2014, 202, 1305–1312.
56. A. F. Pereira, K. U. Kumar, W. F. Silva, W. Q. Santos, D. Jaque and C. Jacinto, Sens. Actuators, B, 2015, 213, 65–71.
57. W. Xu, X. Y. Gao, L. J. Zheng, Z. G. Zhang and W. W. Cao, Sens. Actuators, B, 2012, 173, 250–253.
58. D. Y. Li, Y. X. Wang, X. R. Zhang, K. Yang, L. Liu and Y. L. Song, Opt. Commun., 2012, 285, 1925–1928.
59. V. Lojpur, M. Nikolic, L. Mancic, O. Milosevic and M. D. Dramicanin, Ceram. Int., 2013, 39, 1129–1134.
60. K. Z. Zheng, Z. Y. Liu, C. J. Lv and W. P. Qin, J. Mater. Chem. C, 2013, 1, 5502–5507.
61. X. F. Wang, J. Zheng, Y. Xuan and X. H. Yan, Opt. Express, 2013, 21, 21596–21606.
62. A. K. Singh, Sens. Actuators, A, 2007, 136, 173–177.
63. A. K. Singh and S. B. Rai, Appl. Phys. B, 2007, 86, 661–666.
64. P. Du, L. H. Luo and J. S. Yu, J. Alloys Compd., 2015, 632, 73–77.
65. R. K. Verma and S. B. Rai, J. Quant. Spectrosc. Radiat. Transfer, 2012, 113, 1594–1600.
66. R. Dey, A. Kumari, A. K. Soni and V. K. Rai, Sens. Actuators, B, 2015, 210, 581–588.
67. W. Xu, H. Zhao, Y. X. Li, L. J. Zheng, Z. G. Zhang and W. W. Cao, Sens. Actuators, B, 2013, 188, 1096–1100.
68. W. Xu, X. Y. Gao, L. J. Zheng, Z. G. Zhang and W. W. Cao, Opt. Express, 2012, 20, 18127–18137.
69. C. P. Rodríguez, L. L. Martín, S. F. L. Luis, I. R. Martín, K. K. Kumar and C. K. Jayasankar, Sens. Actuators, B, 2014, 195, 324–331.
70. W. Xu, Q. T. Song, L. J. Zheng, Z. G. Zhang and W. W. Cao, Opt. Lett., 2014, 39, 4635–4638.
71. W. Xu, H. Zhao, Z. G. Zhang and W. W. Cao, Sens. Actuators, B, 2013, 178, 520–524.
72. Z. M. Cao, S. S. Zhou, G. C. Jiang, Y. H. Chen, C. K. Duan and M. Yin, Curr. Appl. Phys., 2014, 14, 1067–1071.
73. Z. Boruc, M. Kaczkan, B. Fetlinski, S. Turczynski and M. Malinowski, Opt. Lett., 2012, 37, 5214–5216.
74. D. Q. Chen, Y. L. Yu, P. Huang, H. Lin, Z. F. Shan and Y. S. Wang, Acta Mater., 2010, 58, 3035–3041.
75. S. H. Park, K. H. Lee, S. Unithrattil, H. S. Yoon, H. G. Jang and W. B. Im, J. Phys. Chem. C, 2012, 116, 26850–26856.
76. J. Zhang, Y. M. Yang, Y. Z. Liu, C. Mi, G. Li, B. Han, Y. Zhang and H. J. Seo, J. Am. Ceram. Soc., 2015, 98, 1567–1573.
77. K. W. Meert, V. A. Morozov, A. M. Abakumov, J. Hadermann, D. Poelman and P. F. Smet, Opt. Express, 2014, 22, 961–972.
78. Y. Tian, B. N. Tian, C. Cui, P. Huang, L. Wang and B. J. Chen, Opt. Lett., 2014, 39, 4164–4167.
79. L. L. Shi, C. Y. Li and Q. Su, Opt. Lett., 2011, 36, 582–584.
80. M. G. Nikolić, Ź. Antić, S. Ćulubrk, J. M. Nedeljković and M. D. Dramićanin, Sens. Actuators, B, 2014, 201, 46–50.

---