Up/down-converted green luminescence of Er³⁺–Yb³⁺ doped paramagnetic gadolinium molybdate: a highly sensitive thermographic phosphor for multifunctional applications

Abstract

A series of Er³⁺–Yb³⁺ doped gadolinium molybdate phosphors were synthesized via hydrothermal method with varying Er³⁺ and Yb³⁺ concentrations and their thermal stability, crystal phase formation, particle morphology and photoluminescence properties were explored. The effects of rare earth doping concentration and annealing temperature on upconversion and downconversion properties have been investigated upon 980 nm and 380 nm light excitation and explained with the variation in lifetime of the ⁴S_3/2 level of Er³⁺. The materials were further investigated to look into the effect of Er³⁺-concentration on optical temperature sensing and nano-heating behavior. Temperature sensing measurements were performed by the fluorescence intensity ratio technique using the transitions from the two thermally coupled energy levels (²H_11/2/⁴S_3/2 → ⁴I_15/2) of Er³⁺. The maximum temperature sensitivity was obtained as 0.0105 K⁻¹ (at 450 K), which is among the highest measured sensitivities for luminescence based thermometers. Moreover, the material shows very high thermal gain due to laser irradiation, resulting in a temperature rise from 364 K to 683 K as the excitation power changes from 7.0 to 65 W cm⁻² and defines the present material as a highly sensitive thermographic phosphor. Additionally, the paramagnetic nature and effect of the magnetic field on upconversion properties of this phosphor have also been explored. The thermally-stable, paramagnetic Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor particles seem to be potential candidates for displays, remote temperature sensing, optical heaters, magneto-optic modulators and bio-imaging applications.

---

1. Introduction

Upconversion (UC) emission, also known as anti-Stokes emission, is a process where low energy photons are converted into high energy photons through sequential absorption by a suitable medium. It offers many conceivable applications in optical sensors, telecommunication, solid state lasers, infrared quantum counters, displays and lightning devices including biomedical applications. The UC emission is generally achieved in lanthanide ions owing to their abundant intra-4f configurational energy levels with several metastable levels.^1–6 Among all rare earth (RE) elements, the erbium (Er³⁺) ion is identified as the most suitable candidate for conversion of infrared to visible light through the UC process. Moreover, the absorption of 980 nm excitation radiation by Er³⁺ ions can be increased several times through the introduction of Yb³⁺ ions,⁷ which enhances the upconversion intensity multiple times.

The colour tuning of UC emission using sensitizer/activator combinations and by adjusting their relative concentrations have been the focus of extensive research for application in display devices.^8–10 Several attempts have been devoted into this area by different groups.^9,10 For instance, Solis et al.⁹ have reported tunability of green to red UC emission by changing the concentration of Er³⁺ and Yb³⁺ ions in ZrO₂ nanocrystals. Song et al.¹⁰ have reported green to yellow colour tunability in Gd₂O₂S: Yb³⁺/Er³⁺ microphosphor by changing the Yb³⁺ concentration. Selection of host is also a way to change the emission colour as different hosts have different lattice environments. Therefore, UC process is strongly influenced by the properties of the host lattice and its interaction with the doped rare earth ions.^11–13

Molybdate compounds as fluorescent medium have attracted recent interest because of their several advantages, e.g. having broad and intense absorption band due to the charge transfer (CT) transition from oxygen to metal ion in the near-UV region, high emission intensity, well dispersibility of RE ions etc.^14,15 He et al.¹⁴ have reported intense red emitting Gd₂(MoO₄)₃: Sm³⁺ phosphor for solid state lighting application. Zhao et al.¹⁵ have investigated red emitting Eu³⁺ doped Gd₂Mo₃O₉ phosphor for white-light-emitting-diode (WLEDs) application. Zhang et al.¹⁶ have discussed the luminescence properties of Dy³⁺ doped Gd₂Mo₃O₉ phosphor prepared via solid-state reaction method. All such investigations on Gd₂Mo₃O₉ phosphor have been carried out for photoluminescence upon ultraviolet (UV) or blue light excitations (doped with Sm³⁺, Eu³⁺, Dy³⁺) but there are only limited studies on the UC properties of Gd₂Mo₃O₉ phosphor.^17,18 Moreover, trivalent gadolinium (Gd³⁺) shows magnetic property and thus makes this phosphor useful for magnetic resonance imaging (MRI).^19,20 The presence of Gd³⁺ ions having seven unpaired 4f electrons which are closely bound to the nucleus and effectively shielded by the outer closed shells 5s²5p⁶ makes it paramagnetic in nature. The studies on the magnetic properties of Gd³⁺ ion based host materials such as GdF₃, GdPO₄ and NaGdF₄ (ref. 19–21) have been reported but the paramagnetic properties of Gd₂Mo₃O₉ phosphor have not yet been explored. Herein, it is interesting to observe the dual emission mode of Gd₂Mo₃O₉: Er³⁺/Yb³⁺, which can convert NIR as well as UV radiation into visible light. Another reason of selection of Er³⁺ ion is its capability of self-referenced temperature sensing, useful for non-contact based thermometry.^22–26 Moreover, at higher excitation power RE doped phosphors show internal heating due to laser light irradiation and demands potential application in photo-thermal therapy, drilling nanoholes in organic and soft materials.^27,28

In this paper, authors have investigated UC properties, temperature sensing and laser induced heat generation in Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor at different rare earth doping concentration and annealing temperature. Authors have also explored the effect of magnetic field on UC emission and paramagnetic nature of the present material.

2. Experimental details

2.1 Material synthesis

The Er³⁺/Yb³⁺ doped Gd₂Mo₃O₉ phosphors were prepared by hydrothermal method using ethylene glycol (EG) as chelating agent. The composition of the sample was as follows:

(100 − x − y) mol% Gd₂Mo₃O₉ + x mol% Er³⁺ + y mol% Yb³⁺

where, x = 0.3, 1, 2, 3, 4 and 5 mol% and y = 1, 2, 3 and 4 mol%.

The starting materials: gadolinium oxide (Gd₂O₃, 99.995% Merck, India), ammonium heptamolybdate tetrahydrate ((NH₄)₆Mo₇O₂₄·4H₂O, 99% Merck, India), erbium oxide (Er₂O₃, 99.99%, Otto, India) and ytterbium oxide (Yb₂O₃, 99.99%, Otto, India) were taken in stoichiometric proportion. The synthesis procedure is described as follows: first the nitrate solutions were prepared by dissolving appropriate amount of Gd₂O₃, Er₂O₃ and Yb₂O₃ in concentrated nitric acid (HNO₃). The excess nitric acid was removed by heating the solution at 80 °C. After that, nitrate solutions were mixed in a beaker by adding 10 ml of de-ionized water under stirring condition at room temperature. In another beaker, (NH₄)₆Mo₇O₂₄·4H₂O was dissolved in 40 ml of de-ionized water under continuous stirring. This solution was then added up drop wise into the former solution and 25 ml of ethylene glycol (EG) was added into as-obtained solution. The pH of the solution was adjusted between 8 and 9 using NaOH under constant stirring condition. After being stirred for half an hour, the resultant milky colloidal solution was transferred into a Teflon-lined stainless steel autoclave having 80 ml capacity and heated at 200 °C for 24 h for hydrothermal treatment. After cooling down to room temperature, the precipitate was collected by centrifugation and washed with de-ionized water and ethanol for several times followed by drying in an oven at 80 °C for 6 h. The powder was further annealed at four different temperatures; 600 °C, 800 °C, 1000 °C and 1200 °C for 3 h.

2.2 Characterizations

Thermogravimetry (TG) and differential thermal analysis (DTA) were carried out on NETZSCH STA 449F3 thermal analyzer in nitrogen atmosphere at a heating rate of 5 °C min⁻¹ in the temperature range 30–1000 °C. The X-ray diffraction (XRD) data were measured on Bruker-D8 Advanced X-ray diffractometer using Cu-K_α (1.5405 Å) radiation. Infrared absorption spectra were recorded on FTIR spectrometer (Perkin Elmer spectrum RXI) using KBr pellet technique in the wavenumber range 4000–400 cm⁻¹. Field emission scanning electron microscopy (FESEM) images were taken on ZEISS Suppra55. Transmission electron microscopy (TEM) images of the prepared samples were inspected on Hitachi (H-7500). The electronic absorption spectra were recorded using Perkin-Elmer, Lambda 950, UV-Visible-NIR spectrophotometer in the wavelength range of 200–1200 nm. The downconversion photoluminescence measurement were carried out on Hitachi fluorescence spectrometer, model F-2500. The UC emission spectra were recorded on SP2300 grating spectrograph (Princeton Instruments, USA) using a 980 nm CW diode laser as excitation source. The lifetime measurements were performed by chopping the CW laser beam at 15 Hz and recorded the decay curves with the help of a digital storage oscilloscope. For temperature dependent UC emission study, a thermocouple was placed close to laser spot on the sample surface and the spectra were recorded on a CCD spectrometer (Model: ULS2048X64, Avantes, USA). The laser beam power was set at 15 W cm⁻² and a chopper was used to chop the CW light to avoid the direct heating of the material. An electromagnet was used to study the effect of magnetic field on UC emission. The magnetic field versus magnetization measurement was conducted using a vibrating sample magnetometer at room temperature and 77 K.

3. Results and discussion

3.1 Thermal analysis

The TG/DTA curve of the as-synthesized phosphor is shown in Fig. 1. The TG curve shows three stages of weight loss. The first stage of TG curve shows a weight loss of ∼5.50%, associated with endothermic peak at 110 °C, and is attributed to the evaporation of absorbed water molecules. The second stage shows the weight loss of 4.92% in the temperature range 350–700 °C, accompanied by an exothermic peak at 350 °C and broad endothermic peak at 700 °C that can be mainly ascribed to the decomposition of nitrate into oxide and the loss of remaining organic components. Furthermore, in the third stage when the temperature exceeds 700 °C, weight loss of 2.14% is observed up to the temperature of 1000 °C accompanied by exothermic peak at 830 °C. The broad exothermic peak at 830 °C is assigned to the crystallization of Gd₂Mo₃O₉ phase.²⁹ Graph indicates that 800 °C temperature is enough to start the crystallization of Gd₂Mo₃O₉ phase.

Fig. 1 Thermogravimetry and differential thermal analysis curve of as-synthesized Gd₂Mo₃O₉: Er³⁺/Yb³⁺.

3.2 X-ray diffraction analysis

The X-ray diffraction patterns of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ annealed at 800 °C, 1000 °C and 1200 °C are shown in Fig. 2(a). All the diffraction patterns are in good agreement with the standard data of JCPDS (file no. 33-0548) confirming formation of pure tetragonal phase of Gd₂Mo₃O₉. At 800 °C annealing temperature, the sample is well crystalline and represents seven prominent peaks as shown in Fig. 2(a). As the sample is annealed at 1000 °C, the peak intensity is found to increase, indicating improvement of crystallinity. However, further annealing at 1200 °C decreases the diffraction peak intensity referring the loss of crystallinity of the sample. Therefore, the optimum annealing temperature is 1000 °C for phase formation of Gd₂Mo₃O₉. The lattice parameters, unit cell volume and lattice strain were calculated and summarized in Table 1 which shows that the values of lattice parameters and unit cell volume decrease with increasing the annealing temperature. Moreover, the diffraction peaks slightly shifts toward the higher 2θ value for the sample annealed at 1200 °C as shown in Fig. 2(b). The shifting of diffraction peak at 1200 °C temperature is due to replacement of the lattice sites of Gd₂Mo₃O₉ by RE ions. The ionic radii of Er³⁺ (0.89 Å) and Yb³⁺ (0.86 Å) are smaller than that of Gd³⁺ (1.05 Å). So, it is expected that Er³⁺ and Yb³⁺ ions substitute the Gd³⁺ ions, causing contraction in unit cell volume.³⁰ The strain present in the lattice was determined from the Williamson–Hall relation³¹where, D is the crystallite size (in nm), β is full width at half maximum, θ is Bragg diffraction angle, λ is the wavelength of radiation (1.54 Å). The strain (ε) is calculated from the slope of β cos θ/λ vs. sin θ/λ plot as shown in Fig. 2(c). The positive slope which indicates tensile strain is observed for the samples annealed at 800 °C and 1000 °C. The slope becomes negative at 1200 °C and tensile strain goes towards compressive strain. The compressive strain is possible when ions of larger radii (Gd³⁺) are replaced by smaller radii (Er³⁺, Yb³⁺).

Fig. 2 (a) X-ray diffraction patterns of 0.3 mol% Er³⁺ and 3.0 mol% Yb³⁺ doped Gd₂Mo₃O₉ annealed at three different temperatures with standard JCPDS data; (b) enlarged XRD patterns of (112) diffraction peak for all three samples; (c) Williamson–Hall plots of the three samples annealed at 800, 1000 and 1200 °C. The peak marked by asterisk in (a) is not assigned.

Table 1 Structural parameters of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor

Gd2Mo3O9: Er^3+/Yb^3+ annealing temperature (°C)	Lattice parameters (Å)	Unit cell volume (V) (Å^3)	Lattice strain (ε)
800	a = b = 5.198 ± 0.012, c = 11.398 ± 0.058	308.054	0.000597
1000	a = b = 5.192 ± 0.002, c = 11.382 ± 0.025	306.886	0.000307
1200	a = b = 5.160 ± 0.016, c = 11.446 ± 0.026	304.793	−0.00469

---

3.3 Fourier transform infrared (FTIR) spectroscopy

Fourier transform infrared spectra of the samples are shown in Fig. 3. The as-synthesized sample shows the presence of organic impurities like O–H, N–O and CO₂. The broad absorption band around 3400 cm⁻¹ is due to O–H stretching vibration of water molecules present in the sample. This band is not present in the annealed samples. The band observed around 2372 cm⁻¹ is due to asymmetric stretching mode of CO₂ (ref. 31) and is found in all samples. There are two possible sources of CO₂: (i) impurity in as-synthesized sample, and (ii) atmospheric CO₂. Since, it is present in all the samples; it is assumed that this band comes from atmospheric CO₂. As-synthesized sample also shows the band around 1652 and 1350 cm⁻¹ due to the O–H bending vibration of H₂O and nitrate NO₃⁻ vibration, respectively. The impurity bands are eliminated in annealed samples. The bands at ∼670 cm⁻¹ are assigned to the lattice vibrations of Gd₂Mo₃O₉.

Fig. 3 Fourier transform infrared spectra of (a) as-synthesized (b) annealed at 800 °C (c) annealed at 1000 °C (d) annealed at 1200 °C samples. No impurity band is observed in the annealed samples.

3.4 Microstructure analysis

The field emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) images were used to examine the surface morphology and size of the synthesized particles. Fig. 4(a) and (b) shows the FESEM images of as-synthesized and sample annealed at 1000 °C, respectively. The particles are very small and most of the particles fall within 50–60 nm range in as-synthesized sample as shown in Fig. 4(e). The particle size distribution was calculated by plotting the histogram with Gaussian fitting. However, for annealed sample the particles get aggregated to form bigger particles. The TEM images of as-synthesized sample are shown in Fig. 4(c) and (d). In these images particles are seen to be connected by grain boundary. The energy dispersive X-ray spectroscopy (EDX) was also analyzed to investigate the chemical composition and purity of the sample as shown in Fig. 4(f). The EDX spectrum confirms the presence of gadolinium (Gd), molybdenum (Mo), oxygen (O), erbium (Er) and ytterbium (Yb) elements in the sample. The calculated weight and atomic percentage of these elements are given in the inset of Fig. 4(f). To further see the distribution of constituent elements, the elemental mapping was done for as-synthesized sample which clearly indicates homogeneous distributions of Gd, Mo and O throughout the sample (Fig. 4(g–i)). The dark hollow spots in the images are due to out of focus at sample surface. However, EPMA technique could not be able to detect the minute presence of Er³⁺ and Yb³⁺ ions.

Fig. 4 FESEM image of (a) as-synthesized sample (b) annealed at 1000 °C sample; TEM image of as-synthesized sample (c) at 100 nm resolution (d) at 20 nm resolution; (e) histogram plot of particle size distribution in as-synthesized sample; (f) EDX spectrum; [g–i] elemental mapping of Gd, O and Mo in Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor.

3.5 UV-visible absorption spectroscopy

The UV-Visible absorption spectrum of 0.3 mol% Er³⁺ and 3.0 mol% Yb³⁺ doped phosphor annealed at 1000 °C was recorded in diffuse reflectance mode against a reference standard of BaSO₄ compound and graph is shown in Fig. 5. In the diffuse reflectance spectrum the bands are observed at 488, 521 and 656 nm due to the ⁴F_7/2 ← ⁴I_15/2, ²H_11/2 ← ⁴I_15/2 and ⁴F_9/2 ← ⁴I_15/2 transitions, respectively of Er³⁺ ion. The band at 978 nm is observed due to the ²F_5/2 ← ²F_7/2 transition of Yb³⁺ ion. The broad absorption band around 300 nm is assigned to the band absorption of the host. The sample is transparent to light in the range of 410–1200 nm.

Fig. 5 The UV-Visible absorption spectrum in diffuse reflectance mode of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ annealed at 1000 °C.

3.6 Effect of Er³⁺ and Yb³⁺ concentrations on UC emission

Fig. 6(a) compares the 980 nm wavelength excited UC emission spectra with normalized green emission of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ at six different concentrations of Er³⁺ with fixed Yb³⁺ concentration (3.0 mol%). The emission bands at 527, 547 and 660 nm wavelengths are assigned to the ²H_11/2 → ⁴I_15/2, ⁴S_3/2 → ⁴I_15/2 and ⁴F_9/2 → ⁴I_15/2 transitions of Er³⁺ ion, respectively. As the Er³⁺ concentration increases the intensity of red emission increases. As a result, at 5 mol% Er³⁺ the intensity of red band dominate over the green band whereas at 0.3 mol% Er³⁺ doped sample the green emission is stronger than the red emission. Although, overall emission intensity decreases with increasing Er³⁺ concentration. The intensity ratio of red to green emission (R/G) as a function of Er³⁺ concentration is plotted in Fig. 6(b). The value of R/G increases monotonically from 0.30 to 2.78 (more than 9 times) as Er³⁺ concentration increases from 0.3 to 5 mol%. The reason behind the increase of R/G is mainly attributed to the increase of cross-relaxation process: Er³⁺ (⁴F_7/2) + Er³⁺ (⁴I_11/2) → Er³⁺ (⁴F_9/2) + Er³⁺ (⁴F_9/2)⁹ which increases the population of red (⁴F_9/2) emitting level at the expense of green (²H_11/2/⁴S_3/2) level population. The inset of Fig. 6(b) shows the variation of green and red emission intensity as a function of Er³⁺ concentration. The visible color of the emission is tuned with Er³⁺ concentration and the color coordinates calculated using the 1931 CIE (Commission Internationale de I'Eclairage)³¹ are shown in Fig. 6(c). The CIE color coordinates (x, y) shift from (0.24, 0.73) to (0.42, 0.52) as Er³⁺ concentration increases from 0.3 to 5 mol% [Table 2]. It reflects that at 0.3 mol% Er³⁺ concentration the phosphor emits green color and with further increase in Er³⁺ concentration the corresponding color changes from green to yellow region due to the enhancement of red emission intensity. The photographs of UC emission upon 980 nm excitation are shown in the insets of Fig. 6(a).

Fig. 6 (a) Green emission normalized comparative UC emission spectra of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ at various concentration of Er³⁺ with fixed Yb³⁺ concentration (3 mol%). (b) Variation of R/G with Er³⁺ concentration. Inset shows variation of green and red emission with Er³⁺ concentration; (c) CIE chromaticity diagram at different Er³⁺ concentrations (d) UC emission spectra of samples at different Yb³⁺ concentrations and at fixed Er³⁺ concentration (0.3 mol%). Inset shows variation of green band (527 nm) with Yb³⁺ concentration; (e) comparison of UC emission spectra for Gd₂Mo₃O₉: 0.3 mol% Er³⁺ and Gd₂Mo₃O₉: 0.3 mol% Er³⁺/3 mol% Yb³⁺ samples.

Table 2 Calculated CIE chromaticity coordinates (x, y) of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ at different Er³⁺ ion concentrations

Samples	CIE chromaticity coordinates
	x	y
Gd2Mo3O9: 0.3% Er^3+/3% Yb^3+	0.24	0.73
Gd2Mo3O9: 1% Er^3+/3% Yb^3+	0.34	0.64
Gd2Mo3O9: 2% Er^3+/3% Yb^3+	0.36	0.62
Gd2Mo3O9: 3% Er^3+/3% Yb^3+	0.38	0.61
Gd2Mo3O9: 4% Er^3+/3% Yb^3+	0.40	0.59
Gd2Mo3O9: 5% Er^3+/3% Yb^3+	0.42	0.52

---

In the next step, the sensitizer (Yb³⁺) concentration was varied from 1 to 4 mol% keeping Er³⁺ concentration fixed at 0.3 mol%. The recorded spectra are shown in Fig. 6(d). The Yb³⁺ addition increases the green emission rapidly and optimizes at 3 mol%. Beyond this concentration, a drastic decrease in intensity is noticed due to concentration quenching. The inset of Fig. 6(d) shows the variation of green emission as a function of Yb³⁺ concentration.

Furthermore, the UC emission spectrum of singly Er³⁺ (0.3 mol%) doped sample is compared with optimized one i.e. Er³⁺ (0.3 mol%)/Yb³⁺ (3 mol%) sample as shown in Fig. 6(e). Overall enhancement of 9 and 6 times is seen for green and red emission bands, respectively. In order to investigate the number of photons involved in the UC processes, the pump power dependence of the green and red emission bands was investigated for Gd₂Mo₃O₉: Er³⁺ and Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphors. At moderate pump powers, the intensity of UC luminescence (I) is proportional to the n-th power of the input laser power (P), i.e. I ∝ Pⁿ, where n is the number of photons involved in the particular UC process.³² Fig. 7(a) and (b) show the ln I versus ln P plots for Gd₂Mo₃O₉: Er³⁺ and Gd₂Mo₃O₉: Er³⁺/Yb³⁺, respectively. For both the samples, the slopes are reduced at high pump powers. The slope, ‘n’ calculated at two pump power regions are presented in Table 3. In moderate power regime the slopes for 527, 547 and 660 nm bands are 1.83, 1.61 and 1.32, respectively for the singly Er³⁺ doped phosphor, whereas in Er³⁺/Yb³⁺ doped phosphor the corresponding slopes get reduced to 1.17, 0.7 and 0.5, respectively. Among these slopes, the values 0.7 (547 nm) and 0.5 (660 nm) are much lower than the expected two photon values. The decrease in slope value from the expected integer numbers in Er³⁺/Yb³⁺ doped sample is due to the increase of energy transfer (ET) from Yb³⁺ to Er³⁺ ions, resulting saturation of intermediate ⁴I_11/2 and ⁴I_13/2 levels. For example, when the energy transfer from Yb³⁺ to Er³⁺ is efficient, the ⁴I_13/2 level gets populated via – Yb³⁺ (²F_5/2) + Er³⁺ (⁴I_15/2) →Yb³⁺ (²F_7/2) + Er³⁺ (⁴I_11/2) followed by non-radiative decay from ⁴I_11/2 level. It is already reported that ET process reduces the slope values from the integer numbers.³³ Despite the low slopes in Er³⁺/Yb³⁺, the intensity is found much higher compared to Er³⁺ doped sample. There are several theoretical models in literature where it has been proved that due to saturation mechanism of intermediate levels the number of photons decrease from the expected value.^32–34 Moreover, in high pump power regime the slopes for all the emission bands decrease which has been explained by a complete theoretical description by our group in previous report on BaTiO₃: Er³⁺/Yb³⁺ phosphor.³³ It is concluded on the basis of previous work that the UC emission in Gd₂Mo₃O₉: Er³⁺ is predominantly governed by excited state absorption (ESA) mechanism whereas in Gd₂Mo₃O₉: Er³⁺/Yb³⁺ it is governed by energy transfer (ET) process. The excitation and emission pathways are shown in energy level diagram, Fig. 7(c).

Fig. 7 Pump power dependence of green and red emission bands in (a) Gd₂Mo₃O₉: Er³⁺ (b) Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphors; (c) energy level diagram of Er³⁺ and Yb³⁺ ions and the proposed UC mechanisms under 980 nm laser light excitation.

Table 3 Slopes of ln P vs. ln I plots

Emission bands	Gd2Mo3O9: Er^3+		Gd2Mo3O9: Er^3+/Yb^3+
	Moderate power	High power	Moderate power	High power
527 nm	1.83	0.8	1.17	0.55
547 nm	1.61	0.4	0.7	0.06
660 nm	1.32	0.2	0.5	0.02

---

3.7 Effect of annealing temperature on upconversion and downconversion (DC) emissions

In order to study the effect of annealing temperature on UC and DC emissions, the as-synthesized (ASP) sample was heat treated at 600 °C, 800 °C, 1000 °C and 1200 °C for 3 h. A comparison of upconversion spectra for the optimized sample is shown in Fig. 8(a). The intensity of both the green and red bands increases with annealing temperature upto 1000 °C. This enhancement is higher for green emission compared to the red emission. In 1000 °C annealed sample, the intensities of green and red bands are 8 and 4 times higher than that of the 600 °C annealed sample. The enhancement in intensity could be explained on the basis of FTIR result. The FTIR spectra show that the organic impurities such as OH⁻ and CO₂ are completely removed from the sample at 1000 °C annealing temperature. Subsequently, the non-radiative decay rates decrease, providing strong UC emission. Apart from this, it can be seen from the FESEM images that the particle size increases at 1000 °C annealing temperature. It may also be the reason to increase the UC intensity. However, UC intensity drastically decreases for the sample annealed at 1200 °C, which could be due to creation of defects in the sample. These defects are created due to removal of oxygen from the sample. A similar feature is also observed in previous reports.^31,35 The defect chemistry could be written as:

Fig. 8 Comparison of Gd₂Mo₃O₉: 0.3 mol% Er³⁺/3 mol% Yb³⁺ emission (a) 980 nm excited UC emission spectra at power density of 15 W cm⁻² for as-synthesized and annealed at 600, 800, 1000 and 1200 °C; (b) 380 nm excitation downconversion emission spectra at four different annealing temperatures. Inset shows the excitation spectrum at λ_exc = 555 nm.

Due to creation of lattice defect, the crystallinity of the sample is supposed to decrease and it was observed from the XRD that the intensity counts for the sample annealed at 1200 °C are lower compared to the sample annealed at 1000 °C. Moreover, the FWHM of the XRD peaks are broadened for sample annealed at 1200 °C. These two observations support the assumption of the oxygen removal from the sample.

Before the photoluminescence (PL) emission measurements, the excitation spectrum was collected for the sample annealed at 1000 °C. The PL excitation spectrum is shown in the inset of Fig. 8(b). The sharp bands in the excitation spectrum are due to absorption transitions of Er³⁺. The broad band around 300 nm is assigned to O²⁻ → Mo⁶⁺ charge transfer (CT) band in which electrons are excited from the 2p orbitals of oxygen (top of the valence band) to the 4d orbitals of molybdate (bottom of the conduction band). Prominent absorption bands at 380, 407, 451, 470 and 488 nm are ascribed to the ⁴G_11/2 ← ⁴I_15/2, ²H_9/2 ← ⁴I_15/2, ⁴F_5/2 ← ⁴I_15/2, ⁴F_3/2 ← ⁴I_15/2 and ⁴F_7/2 ← ⁴I_15/2 transitions, respectively indicating efficient excitation in near-UV (380 nm) and blue (400–490 nm) regions in Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor. Corresponding to intense absorption at 380 nm, the excitation wavelength was selected and the recorded emission spectra are compared with UC spectra as shown in Fig. 8(a). Similar to UC pattern, the DC emission intensity is also observed to increase with annealing temperature upto 1000 °C and thereafter decreases. Also it is noted that in DC pattern the emission intensity difference between 1000 °C and 1200 °C annealed samples is low as compared to UC case and leads us to conclude that defects play a little role in downconversion emission. The spectra show strong green emission at 530 nm and 555 nm ascribed to ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 transitions of Er³⁺. Interestingly, there is no emission corresponding to ⁴F_9/2 → ⁴I_15/2 transition. This reflects the absence of non-radiative transition from the ²H_11/2(⁴S_3/2) to ⁴F_9/2 level.

Emission decay measurements were performed under 980 nm excitation for the ⁴S_3/2 level of Er³⁺ to see the effect of annealing temperature and defects. Fig. 9(a–d) shows the decay curves for ⁴S_3/2 → ⁴I_15/2 transition at four different annealing temperatures: 600 °C, 800 °C, 1000 °C and 1200 °C. All the decay curves were well fitted to single exponential formula, expressed as:

where, I(t) and I(0) denote emission intensities at time t and 0, respectively; t is the time and τ is the decay time of the emitting level. The measured effective decay time for ⁴S_3/2 emitting level at different annealing temperatures are mentioned in the figure. The lifetime increases with annealing temperature and is maximum for the sample annealed at 1000 °C. The lifetime decreases at 1200 °C annealing temperature. This observation again correlates the defects creation in the sample.

Fig. 9 Fluorescence lifetime curves of ⁴S_3/2 level of Er³⁺ under 980 nm excitation with single exponential fitting for Gd₂Mo₃O₉: 0.3 mol% Er³⁺/3 mol% Yb³⁺ phosphor annealed at (a) 600 °C (b) 800 °C (c) 1000 °C (d) 1200 °C.

Based on above observations, the mechanisms of the excitation and emission processes are schematically shown in the energy level diagram in Fig. 7(c). The two UC mechanisms are found responsible for intense green and red emission: (1) the excited state absorption (ESA) in Er³⁺ and (2) energy transfer upconversion (ETU) from Yb³⁺ to Er³⁺ ions. In singly Er³⁺ doped phosphor ESA is the dominating mechanism as slopes in pump power studies are above the value 1.5. In Er³⁺/Yb³⁺ co-doped phosphor, ET dominates over ESA and results large enhancement in UC intensity. The full description of the mechanism can be found in our earlier report.³¹ Based on the comparative spectra of UC and DC, it is important to discuss that red emission is absent in DC compared to UC process. In UC process the ⁴F_9/2 level is populated through absorption of excitation photon by the ⁴I_13/2 level or energy transfer process: Yb³⁺ (²F_5/2) + Er³⁺ (⁴I_13/2) → Yb³⁺ (²F_7/2) + Er³⁺ (⁴F_9/2). In DC, the only possible way of populating the ⁴F_9/2 level is the non-radiative transition from ⁴S_3/2 to ⁴F_9/2 level. But this channel is absent in DC process resulting no red emission.

3.8 Temperature sensing and optical heating

To investigate the temperature sensing ability of the Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor and to observe the effect of doping concentration on the sensing efficiency, the temperature dependent UC measurements were carried out. In Fig. 10(a–d), a comparison of intensity variation with temperature is shown for the four different concentrations of Er³⁺ (0.3, 1, 2 and 3 mol%). For clarity, comparison is shown only for three temperatures namely 300 K, 390 K and 460 K at a fixed excitation power of 15 W cm⁻². The emission intensities of both the green emitting bands decrease with increasing temperature but the rate of decrease of 549 nm band is faster than the 524 nm band for all the samples. Fig. 10(e–h) show the variation of fluorescence intensity ratio (FIR) of the emission bands centered at 524 nm and 549 nm as a function of temperature. The FIR values increase with temperature for all the samples but rate of increment in FIR is more prominent for the sample containing lowest Er³⁺ concentration (0.3 mol%). The rate decreases with increase in Er³⁺ doping concentration. The fluorescence intensity ratio (FIR) of two thermally coupled levels, ²H_11/2 and ⁴S_3/2 of Er³⁺ follows Boltzmann-type population distribution and can be written using the following formula:²⁴where, I₅₂₄ and I₅₄₉ are the integrated intensities corresponding to the ²H_11/2 → ⁴I_15/2 (524 nm) and ⁴S_3/2 → ⁴I_15/2 (549 nm) transitions, respectively. W_H and W_S are the radiative probabilities of the transitions, g_H and g_S are the (2J + 1) degeneracy of levels ²H_11/2 and ⁴S_3/2 respectively and hν_H and hν_S are the photon energies of the ²H_11/2 → ⁴I_15/2 and ⁴S_3/2 → ⁴I_15/2 transitions respectively, ΔE is the energy gap between the two emitting levels, k_B is the Boltzmann constant and T is the absolute temperature. Eqn (3) can be expressed as follows:

Fig. 10 (a–d) Temperature dependent UC emission spectra for 524 nm (²H_11/2 → ⁴I_15/2) and 549 nm (⁴S_3/2 → ⁴I_15/2) bands of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor excited by 980 nm excitation for different concentration of Er³⁺ (0.3 to 3 mol%); (e–h) variation of FIR as a function of absolute temperature for all the four phosphors.

The plot of ln(FIR) as a function of inverse absolute temperature in the range of 300–460 K for 0.3, 1.0, 2.0 and 3.0 mol% concentrations of Er³⁺ are shown in Fig. 11(a–d). The experimental data is well fitted to straight line with the slopes (ΔE/k) of about – 1255, 768, 734 and 723 cm⁻¹ for 0.3, 1.0, 2.0 and 3.0 mol% doped samples and corresponding calculated values of ΔE from the graphs are obtained as – 872, 534, 510 and 502 cm⁻¹. In the spectra shown in Fig. 10(a), the relative intensities of Stark components of ²H_11/2 and ⁴S_3/2 levels vary with Er³⁺ concentrations and hence it may leads to vary the effective energy gap with Er³⁺ concentration. Two additional heating–cooling cycles (two heating + two cooling) were performed and a good repeatability was obtained (data are provided in ESI Fig. 1†). The experimentally observed energy gaps are used to calculate the sensor sensitivity. The values of co-efficient B were obtained according to the fitting curves of the experimental data.

Fig. 11 (a–d) The monolog plot of the FIR (I₅₂₄/I₅₄₉) as a function of inverse absolute temperature for different concentrations of Er³⁺ (0.3 to 3 mol%) (e–h) variation of absolute temperature sensitivity (S) as a function of temperature within 300–480 K.

The sensor sensitivity is an important parameter for temperature sensing and is defined as the rate at which fluorescence intensity ratio changes with temperature. The absolute sensitivity (S) is expressed by the following equation:²⁴

here, ΔE is experimentally calculated value from the FIR graph. This equation shows that at a given temperature, higher value of ΔE would give higher sensitivity. The sensor sensitivity was also evaluated for all the samples (i.e. for 0.3, 1, 2 and 3 mol% Er³⁺) as a function of temperature and plots are shown in Fig. 11(e–h). Fig. 11(e–h) shows that Er³⁺ concentration strongly affects the sensitivity of the material. Moreover, the position of maximum sensitivity shifts towards lower temperatures with the increase in Er³⁺ concentration. The maximum value of sensitivity is found as 10.57 × 10⁻³ K⁻¹ at 450 K temperature for 0.3 mol% Er³⁺ doped sample while for 3 mol% Er³⁺ doping the maximum sensitivity reduces to 3.56 × 10⁻³ K⁻¹ at 373 K temperature. The decrease of sensitivity with increasing Er³⁺ concentration may be correlated with the increase of energy transfer between the Er³⁺–Er³⁺ ion through cross relaxation process Er³⁺ (⁴F_7/2) + Er³⁺ (⁴I_11/2) → Er³⁺ (⁴F_9/2) + Er³⁺ (⁴F_9/2) which depopulates both ²H_11/2 and ⁴S_3/2 levels leading to the reduction in the green emission intensity and hence in the FIR values. This energy transfer process increases with Er³⁺ concentration and sensitivity goes down continuously with concentration. Table 4 presents the comparison of sensor sensitivity and their temperature range with similar studies on different materials.^{24–26,33,36–40} It is noticed that sensitivity obtained for the present Gd₂Mo₃O₉ phosphor is better than several reported works. Thus, Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor seems to be an interesting host material for temperature sensing application with high sensor sensitivity.

Table 4 The comparison of maximum value of temperature sensitivity in Er³⁺ based materials by FIR technique

Materials	Temperature range	Maximum sensitivity	References
Er–Yb: YVO4	300–485 K	0.0116 K^−1 at 380 K	24
Er–Yb–Mo: Al2O3	294–973 K	0.0051 K^−1 at 443 K	25
Er–Yb: NaGd(WO4)2	293–573 K	0.0119 K^−1 at 453K	26
Er–Yb–Zn: BaTiO3	120–505 K	0.0047 K^−1 at 430 K	33
Er: PKAZLF glass	298–773 K	0.0079 K^−1 at 630 K	36
Er–Yb: Silicate glass	296–723 K	0.0033 K^−1 at 296 K	37
Er–Yb: YNbO4	298–673 K	0.0072 K^−1 at 406K	38
Er–Yb–Gd2O3	298–723 K	0.0084 K^−1 at 570K	39
Er: Sr2YbF7 glass–ceramic	300–500 K	0.0062 K^−1 at 500K	40
Er–Yb: Gd2Mo3O9	300–460 K	0.0105 K^−1 at 450 K	This work (0.3 mol% Er^3+)

---

Furthermore, in order to examine the influence of Er³⁺ ion concentration on optical heating of the samples caused by the laser excitation (980 nm) and emission processes, the laser beam was focused on the sample surface within 3 mm² area. Pump power was increased and FIR was calculated with excitation power for all above studied samples. The FIR vs. pump power plot is shown in Fig. 12(a). It is observed that on increasing the power density from 7 to 65 W cm⁻² the FIR (I₅₂₄/I₅₄₉) value increases linearly for all the samples but the variation of FIR is the fastest for sample having 0.3 mol% Er³⁺ concentration. The same pattern is observed in case of FIR vs. temperature graph [Fig. 10(e–h)]. The increase in FIR is possible when sample temperature increases and it is concluded that laser excitation generates heat in the sample. The FIR values obtained at different pump power can efficiently be correlated with FIR vs. temperature graph. The temperature-gained by the sample particles due to laser excitation for various Er³⁺ concentrations were calculated using the following equation:⁴¹

where, all the terms have their usual meanings. The sample temperature due to laser induced heat for various Er³⁺ concentrations was derived using the values of ΔE and B calculated from the temperature sensing study. The calculated sample temperatures are listed in Table 5. Dependence of sample temperature on excitation power density was also plotted and is shown in Fig. 12(b). The induced temperature is found to increase with increasing the pump power for all the samples. Moreover, the heat generation is the maximum for the lowest Er³⁺ (0.3 mol%) doping and its value decreases as concentration increases. The heat in the sample arises due to involvement of non-radiative transitions. As concentration increases the energy migration from one Er³⁺ to another Er³⁺ ion increases and generates less heat in highly doped samples. This observation again supports energy migration with increasing Er³⁺ concentration. The huge amount of heat generation in the present material is an exceptional achievement for optical heaters investigated so far.²⁴

Fig. 12 (a) Fluorescence intensity ratio of emission bands at 524 nm and 549 nm as a function of pump power density (b) temperature variation of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor as a function of excitation power density upon 980 nm excitation.

Table 5 Calculated values of temperature at various pump power densities for different concentration of Er³⁺ ions

Power density (W cm^−2)	0.3 mol% Er^3+		1 mol% Er^3+		2 mol% Er^3+		3 mol% Er^3+
	FIR	Temp (K)	FIR	Temp (K)	FIR	Temp (K)	FIR	Temp (K)
7	0.87	364	0.65	325	0.50	296	0.40	281
13	1.21	401	0.70	335	0.57	308	0.44	295
19	1.57	438	0.75	347	0.63	321	0.49	308
26	2.00	479	0.81	359	0.70	336	0.54	322
34	2.45	518	0.89	375	0.77	353	0.60	337
40	2.95	562	0.95	388	0.85	367	0.64	352
50	3.49	607	1.03	405	0.93	387	0.71	368
58	3.95	645	1.11	420	1.01	403	0.77	384
65	4.40	683	1.17	433	1.07	417	0.82	396

---

3.9 Magnetic property

The simultaneous existence of luminescence and magnetic properties in a single material is rarely available and are suitable for magnetic resonance imaging, magnetic field detection and biological sensing and detection. Therefore, these materials are of special attention for developing multifunctional materials. In order to know the magnetic nature of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor, the magnetization as a function of applied magnetic field at 293 K and 77 K temperatures were measured and recorded magnetization curves are shown in Fig. 13(a). The material exhibits paramagnetic behavior at both the temperatures, unlike the Gd³⁺ ions exhibit ferromagnetic behavior below room temperature. The straight line crossing the origin verifies the paramagnetic nature. The magnetic moments associated with the Gd³⁺ ions are all localized and non-interacting, giving rise to paramagnetism.^19–21 The magnetic mass susceptibility at 293 K and 77 K were determined to be 1.38 × 10⁻⁵ emu g⁻¹ Oe⁻¹ and 4.86 × 10⁻⁵ emu g⁻¹ Oe⁻¹, respectively. Here, it is observed that the value of magnetic susceptibility increases at low temperature (77 K) due to the reduction in thermal fluctuation, which is a typical behavior in paramagnetic materials described by the Curie's law.¹⁹ The saturation magnetization value is found to be 0.27 emu g⁻¹ at room temperature and 0.73 emu g⁻¹ at 77 K. These obtained results are comparable to the previously reported values.^19,42,43 However, the magnetic nature of the present phosphor broadens an extra feature to this material from application point of view.

Fig. 13 (a) Magnetization versus applied magnetic field in as-synthesized Gd₂Mo₃O₉: Er³⁺/Yb³⁺ sample (b) magnetic field dependent UC emission spectra of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ excited by 980 nm excitation.

We know that the degeneracy of rare earth energy levels in crystalline host materials is removed through the splitting of the energy levels by the crystal field of the host matrix. These energy levels are further split in the presence of weak magnetic field, called Zeeman effect. To know the effect of magnetic field on UC emission, the material was investigated upon 980 nm excitation under magnetic field. Fig. 13(b) compares the UC spectra at different magnetic fields (from 0 to 1 T) having fixed excitation power density (15 W cm⁻²). It is found that intensity of both the green and red emission bands decrease with increasing magnetic field. In the inset of Fig. 13(b), enlarged portion clearly shows the change in UC emission intensity with magnetic field. Due to application of magnetic field, the energy levels split and disturb the resonant excitation process of the system. Thus the population of the UC emitting levels decreases with increasing magnetic field resulting decrease in UC emission intensity.^21,44,45 Similar findings were reported in rare earth doped GdNbO₄,⁴⁵ Gd₂O₃ (ref. 44) and NaGdF₄ (ref. 21) hosts.

4. Conclusions

In conclusion, the luminescence emission in tetragonal phase of Gd₂Mo₃O₉: Er³⁺/Yb³⁺ paramagnetic phosphor, prepared via hydrothermal method was optimized through concentration variation of Er³⁺ and Yb³⁺ ions. The enhancement in upconversion emission is noticed to increase with annealing temperature upto 1000 °C due to removal of organic impurities from the sample. However, a decrease in emission intensity in the sample annealed at 1200 °C is identified due to oxygen deficiency. The material exhibits green to yellow color tunability with Er³⁺ doping concentration. The FIR based temperature sensing (300–460 K) study using the thermally coupled energy levels (²H_11/2 and ⁴S_3/2 of Er³⁺) indicates an important role of Er³⁺ concentration in determining sensor sensitivity. The temperature sensitivity decreases continuously with increasing Er³⁺ ion concentration and the maximum sensitivity is found at 0.3 mol% Er³⁺ and 3.0 mol% Yb³⁺ doping concentration. Moreover, laser induced heat is measured in the sample and the maximum temperature of the sample particles was calculated as 683 K at 65 W cm⁻² excitation power, pointing out huge amount of heat generation. The application of magnetic field on the sample has resulted decrease in upconversion emission intensity. The present Gd₂Mo₃O₉: Er³⁺/Yb³⁺ phosphor is an interesting thermographic phosphor, offering good green up/down-converted emission and could be considered for various applications including temperature sensing, photo-thermal therapy and magnetic field sensors.

Acknowledgements

Authors are thankful to Dr V. K. Rai, and Dr R. Thangavel, Department of Applied Physics, IIT(ISM), Dhanbad for providing us to use their experimental facilities. Authors also acknowledge Prof. S. B. Rai, Department of Physics, BHU, Varanasi for extension of lifetime measurement facility. One of the authors, Ms S. Sinha is thankful to IIT(ISM), Dhanbad, for providing research fellowship. Dr K. Kumar acknowledges Council of Scientific & Industrial Research, New Delhi (project no. 03(1303)/13/EMR-II) for financial assistance.

References

1. Z. Gu, L. Yan, G. Tian, S. Li, Z. Chai and Y. Zhao, Adv. Mater., 2013, 25, 3758.
2. F. Auzel, Chem. Rev., 2004, 104, 139.
3. F. Vetrone, R. Naccache, A. Zamarron, A. Juarranz de la Fuente, F. S. Rodriguez, L. M. Maestro, E. M. Rodriguez, D. Jaque, J. G. Sole and J. A. Capobianco, ACS Nano, 2010, 4, 3254.
4. F. Wang and X. Liu, Chem. Soc. Rev., 2009, 38, 976.
5. G. Wang, Q. Peng and Y. Li, J. Am. Chem. Soc., 2009, 131, 14200.
6. C. Mi, J. Wu, Y. Yang, B. Han and J. Wei, Sci. Rep., 2016, 6, 22545.
7. H. Guo, N. Dong, M. Yin, W. Zhang, L. Lou and S. Xia, J. Phys. Chem. B, 2004, 108, 19205.
8. V. Mahalingam, C. Hazra, R. Naccache, F. Vetrone and J. A. Capobianco, J. Mater. Chem. C, 2013, 1, 6536.
9. D. Solis, E. De la Rosa, O. Meza, L. A. Diaz-Torres and P. Salas, J. Appl. Phys., 2010, 108, 023103.
10. Y. Song, Y. Huang, L. Zhang, Y. Zheng, N. Guo and H. You, RSC Adv., 2012, 2, 4777.
11. S. K. Singh, K. Kumar and S. B. Rai, Appl. Phys. B, 2009, 94, 165.
12. A. K. Singh, S. Singh, D. Kumar, D. K. Rai, S. B. Rai and K. Kumar, Opt. Lett., 2012, 37, 776.
13. S. K. Singh, K. Kumar and S. B. Rai, Appl. Phys. B, 2010, 100, 443.
14. X. He, J. Zhou, N. Lian, J. Sun and M. Guan, J. Lumin., 2010, 130, 743.
15. X. Zhao, X. Wang, B. Chen, Q. Meng, W. Di, G. Ren and Y. Yang, J. Alloys Compd., 2007, 433, 352.
16. L. Zhang, H. Zhong, X. Li, L. Cheng, L. Yao, J. Sun, J. Zhang, R. Hua and B. Chen, Phys. B, 2012, 407, 68.
17. Y. X. Pan and Q. Y. Zhang, Mater. Sci. Eng., B, 2007, 90, 138.
18. J. Sun, B. Xue and H. Du, Mater. Sci. Eng., B, 2013, 178, 822.
19. H. T. Wong, H. L. W. Chan and J. H. Hao, Appl. Phys. Lett., 2009, 95, 022512.
20. T. Grzyb, A. Gruszeczka, R. J. Wiglusz, Z. Sniadecki, B. Idzikowski and S. Lis, J. Mater. Chem., 2012, 22, 22989.
21. Y. Zhang, Q. Xiao, H. He, J. Zhang, G. Dong, J. Han and J. Qiu, J. Mater. Chem. C, 2015, 3, 10140.
22. A. K. Singh, S. K. Singh, B. K. Gupta, R. Prakash and S. B. Rai, Dalton Trans., 2013, 42, 1065.
23. R. Dey and V. K. Rai, Dalton Trans., 2014, 43, 111.
24. M. K. Mahata, K. Kumar and V. K. Rai, Sens. Actuators, B, 2015, 209, 775.
25. B. Dong, B. S. Cao, Z. Q. Feng, X. J. Wang and Y. Y. He, Sens. Actuators, B, 2012, 165, 34.
26. J. Liao, L. Nie, Q. Wang, S. Liu, H. R. Wen and J. Wu, RSC Adv., 2016, 6, 35152.
27. R. K. Verma and S. B. Rai, J. Quant. Spectrosc. Radiat. Transfer, 2012, 113, 1594.
28. V. K. Tikhomirov, K. Driesen, V. D. Rodriguez, P. Gredin, M. Mortier and V. V. Moshchalkov, Opt. Express, 2009, 17, 11794.
29. C. Guo, W. Zhang, L. Luan, T. Chen, H. Cheng and D. Huang, Sens. Actuators, B, 2008, 133, 33.
30. A. Boukerika and L. Guerbous, J. Lumin., 2014, 145, 148.
31. M. K. Mahata, K. Kumar and V. K. Rai, Spectrochim. Acta, Part A, 2014, 124, 285.
32. M. Pollnau, D. R. Gamelin, S. R. Luthi, H. U. Gudel and M. P. Hehlen, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61, 3337.
33. M. K. Mahata, T. Koppe, T. Mondal, C. Brüsewitz, K. Kumar, V. K. Rai, H. Hofsass and U. Vetter, Phys. Chem. Chem. Phys., 2015, 17, 20741.
34. C. Mi, J. Wu, Y. Yang, B. Han and J. Wei, Sci. Rep., 2016, 6, 22545.
35. A. K. Singh, K. Kumar, A. C. Pandey, O. Parkash, S. B. Rai and D. Kumar, Appl. Phys. B, 2011, 104, 1035.
36. N. Vijaya, P. Babu, V. Venkatramu, C. K. Jayasankar, S. F. León-Luis, U. R. Rodríguez-Mendoza, I. R. Martín and V. Lavín, Sens. Actuators, B, 2013, 186, 156.
37. L. Chengren, B. Dong, L. Shufeng and C. Song, Chem. Phys. Lett., 2007, 43, 426.
38. Y. Tian, Y. Tiana, P. Huang, L. Wang, Q. Shi and C. Cui, Chem. Eng. J., 2016, 297, 26.
39. Y. Tian, B. Tian, C. Cui, P. Huang, L. Wang and B. Chen, RSC Adv., 2015, 5, 14123.
40. X. Li, J. K. Cao, Y. L. Wei, Z. R. Yang and H. Guo, J. Am. Ceram. Soc., 2015, 98, 3824.
41. A. Pandey, V. K. Rai, V. Kumar, V. Kumar and H. C. Swart, Sens. Actuators, B, 2015, 209, 352.
42. Z. L. Wang, J. H. Hao and H. L. W. Chan, J. Mater. Chem., 2010, 20, 3178.
43. L. W. Yang, Y. Y. Zhang, J. J. Li, Y. Li, J. X. Zhong and P. K. Chu, Nanoscale, 2010, 2, 2805.
44. S. K. Singh, K. Kumar, M. K. Srivastava, D. K. Rai and S. B. Rai, Opt. Lett., 2010, 35, 1575.
45. A. Dwivedi, A. K. Singh and S. B. Rai, Dalton Trans., 2014, 43, 15906.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20332a

---