Judd–Ofelt parameters and X-ray irradiation results of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors synthesized via a molten salt method

Abstract

Trivalent Eu-activated MNb₂O₆ (M = Sr, Cd, Ni) ceramic phosphors were produced using the molten salt route, which involves a low sintering temperature and provides improved homogeneity. The photoluminescence (PL) and radioluminescence (RL) spectra of phosphors exhibited characteristic Eu³⁺ emissions with ⁵F₀ → ⁷F_j transitions, and strong peaks occurred at the ⁵D₀ → ⁷F₂ transition. The PL and RL emissions of SrNb₂O₆:Eu³⁺ decreased over 3 mol%, while both emissions for CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ increased with increasing Eu³⁺ concentration. The spectral properties of phosphors were evaluated by determining Judd–Ofelt intensity parameters (Ω₂, Ω₄) from the PL emission spectrum. The quantum efficiencies (η_QE%) of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors with the highest emission were found as 61.87%, 41.89%, and 11.87% respectively. Bandwidths (σ_e × Δλ_eff) and optical gains (σ_e × τ) of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors with highest emissions were found as follows; 24.182 × 10⁻²⁸, 28.674 × 10⁻²⁸, 38.647 × 10⁻²⁸ cm³ and 20.441 × 10⁻²⁵, 13.790 × 10⁻²⁵, 3.987 × 10⁻²⁵ cm² s, respectively, corresponding to the ⁵D₀ → ⁷F₂ transition.

---

1. Introduction

The scientific and technological applications of rare earth (RE) elements have seen significant progress in last few decades due to their unique spectroscopic properties in different host lattices, where their superior chemical and thermal stability are along with their outstanding luminescent yield and color purity. Today, trivalent RE ions have a wide range of applications, such as high-performance luminescent devices, magnets, catalysts, and other functional materials.^1–7 When the host lattice is activated by RE³⁺ ions, besides PL, they can cause changes in RL features. In the radioluminescence mechanism, high-energy ionizing radiation (X-rays, alpha particles, and electrons) instantly transforms into thousands of low-energy photons (1–6 eV). The RL feature is of great importance in many fields, such as high-tech industrial applications, homeland security, radiography, biochemical applications, medical devices, biological safety, particle physics, astrophysics, and nuclear material applications.^6,7

The oxide compounds of columbite niobates can be expressed as MNb₂O₆ (M = Sr, Cd, Ni, Co, Mg, Mn, etc.). The symmetry of MNb₂O₆ easily accepts guest ions of a similar ionic size as the metal and Nb ions of the structure. As the MNb₂O₆ structure, the strontium niobate (SrNb₂O₆) compound composed of the SrO₆–NbO₆ octahedral^8,9 stands out due to its luminescence,^7–9 photocatalytic,^10,11 and dielectric¹² properties. The high photocatalytic property of SrNb₂O₆ is attributed to its increasing absorption by the change of its crystal structure.¹⁰ The molten salt method, one of the methods used in the synthesis of niobates has a favorite feature due to some advantages, which provides a low sintering temperature, short reaction time, good composition, and improved homogeneity.^7,8 The advantage of the molten salt method can be evaluated by comparing it with the solid-state reaction method. Here, SrNb₂O₆ was produced by the solid-state reaction method at 1250 °C for 7 h,⁹ whereas SrNb₂O₆ was fabricated using the molten salt method at 650 °C for 4 h.^7,8

In this study, the structural, photoluminescence, and X-ray induced characterizations of SrNb₂O₆:Eu³⁺ were carried out by XRD, SEM, PL, and RL analyses. In addition, the previously reported phosphors of CdNb₂O₆:Eu³⁺ (ref. 13) and NiNb₂O₆:Eu³⁺ (ref. 14) were included in the study, and the Judd–Ofelt intensity parameters and radioluminescence results of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors were reported by comparison with PL and RL analyses. Also, the grain structures of three compounds were examined by SEM analysis.

2 Experimental

In the synthesis of (Sr, Cd, Ni)Nb₂O₆:xEu³⁺ (x% = 0.5, 3, 6 mol%) and undoped ceramic powders by the molten salt method, as starting materials for each synthesis, strontium nitrate (Sr(NO₃)₂) (Sigma-Aldrich, 99%), cadmium nitrate tetrahydrate (Cd(NO₃)₂·4H₂O) (Sigma-Aldrich, 98.5%), nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O) (Sigma-Aldrich, 99%), pure niobium oxide (Nb₂O₅) (Alpha Aesar, 99.9%), and Eu₂O₃ (Alpha Aesar, 99.9%) powder were used in the calculated stoichiometric amounts. The x value was taken as “mol”, which means 2 Eu atoms due to the Eu₂O₃ compound. The syntheses of (Cd,Ni)Nb₂O₆:Eu³⁺ (ref. 13 and ¹⁴) and SrNb₂O₆:Nd³⁺, Dy³⁺ (ref. 7 and ⁸) were reported as intermediate solid solutions. However, the Sr(NO₃)₂ amount was reduced slightly to prevent secondary phase formation in the synthesis of SrNb₂O₆:Nd³⁺, Dy³⁺,^7,8 similarly, 0.98 and 0.95 mol were used instead of 1 mol Sr(NO₃)₂ for the 3 and 6 mol% Eu³⁺ samples, respectively, corresponding to 2 mol of Nb₂O₅. The stoichiometric amounts of the molten salt mixtures were added to the reactant mixtures. For this purpose, the molten salts to be used and their mixing ratios were, respectively, for strontium niobate synthesis, NaCl : KCl, salt : salt molar ratio 6 : 1 and salt : oxide weight ratio 1 : 1, for cadmium niobate synthesis, Li₂SO₄ : Na₂SO₄, salt : salt molar ratio 1 : 1 and salt : oxide weight ratio 1 : 1, for nickel niobate synthesis, Li₂SO₄ : Na₂SO₄, salt : salt molar ratio 1 : 1 and salt : oxide weight ratio 1 : 1. The starting materials for each compound group were mixed to produce the phosphor powders by the molten salt synthesis method and finely ground in an agate mortar to ensure homogeneity. After that, europium oxide (Eu₂O₃) dopant was added. The final powder mixture was thoroughly mixed in an agate mortar for a final time and re-ground to ensure greater homogeneity. Then, mixtures prepared by weighing for each compound were placed and calcined in alumina crucibles at 650 °C for 4 h for strontium niobate, or at 900 °C for 4 h for cadmium niobate and nickel niobate. Phosphor powders obtained as a result of the heat treatment were washed many times with ultrapure water to remove the ionic salts they contained and vacuum filtered several times. The presence of Cl-ions in the filtrate was checked using qualitative reactions.

The powder samples analyses were carried out by XRD (X-ray diffractometer; D2 PHASER, Bruker Corp., Germany) using Cu-K_α (1.54 Å) radiation in between 2θ = 5–70 °C with a scan speed of 2 °C min⁻¹. SEM micrographs of the samples were taken by scanning electron microscopy (SEM; Inspect S50, FEI Corp., USA) after Au coating. Elemental compositions were determined by energy dispersive spectroscopy (EDS; INCAx-Sight 7274 model; 133 eV resolution 5.9 keV, OXFORD Instruments, England) which was attached to the SEM system (JSM-5910LV, JEOL Corp., Japan). EDS was performed at 20 kV SEM acceleration voltage.

PL (photoluminescence) measurements were recorded on an Edinburgh Instruments (UK) FLSP920 model fluorescence spectrometer with xenon lamp (450 W) continuous wavelength lamps for steady-state measurements. Decay times were taken by μF1 and μF2: 5 W or 60 W pulsed xenon microsecond flashlamps producing short microsecond pulses for decay measurements. Decay curves data (B₁, τ₁, B₂, τ₂) were recorded with a time-correlated single-photon counting (TCSPC) system. Radioluminescence (RL) results were obtained using an Horiba Jobin Yvon spectrometer (Japan), which included a liquid nitrogen-cooled CCD detector and an X-ray irradiation source Machlett OEG-50A. The data for the RL were recorded with a detector integration time of 5 s between 200–1000 nm and 2 mm was chosen for the slit widths (input and output). The operating level of the spectrometer was 30 kV and 15 mA (30 Gy min⁻¹) and it was operated in a dark environment and at room temperature.

3. Results and discussions

3.1 Structural analyses

Fig. 1 shows the X-ray diffraction results of undoped SrNb₂O₆ and Eu³⁺-doped SrNb₂O₆ samples. The XRD pattern (JCPDS no: 28-1243) of the samples was indexed by orthorhombic columbite symmetry with the space group Pbcn60. The CdNb₂O₆:Eu³⁺ (ref. 13) and NiNb₂O₆:Eu³⁺ (ref. 14) structures also had the same symmetry, and their symmetry continued up to 6 mol%, while secondary phases were found at low amounts. The XRD-SEM results of SrNb₂O₆:Eu³⁺ were investigated in this study, the detailed X-ray diffraction results of the CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ have been reported in ref. 13 and ¹⁴. The cell parameters of orthorhombic SrNb₂O₆ are a = 5.592 Å, b = 7.722 Å, c = 10.989 Å, and α = β = γ = 90°.⁷ On the other hand, it has been reported that the orthorhombic columbite symmetry and monoclinic symmetry almost fully overlap with each other.^15,16 Accordingly, XRD reflections of SrNb₂O₆ could be indexed with monoclinic symmetry (space group P2₁/c) which exactly fitted with the orthorhombic symmetry.^15,16 Based on the monoclinic symmetry, the cell parameters of SrNb₂O₆ are a = 5.594 Å, b = 7.722 Å, c = 10.986 Å, and α = 90.37°, β = γ = 90°.^17,18

Fig. 1 X-ray diffraction results for undoped SrNb₂O₆ and Eu³⁺-doped SrNb₂O₆ powders.

The SrNb₂O₆ structure composed of corner-shared and edge-shared NbO₆ octahedral aligned along the a-axis and c-axis, respectively, as shown in Fig. 2. The Sr²⁺ and Nb⁵⁺ cations are surrounded by O²⁻ with 8 and 6 CN (coordination number), and their ionic radii (r) are 1.26 Å and 0.64 Å, respectively.^7,9 There is a cationic site with Wyckoff position 4e for the compatibility of the activator ions within the crystal structure.⁹ The Eu³⁺ ions will occupy the SrNb₂O₆ structure with the ionic radii of 0.947 Å for 6 CN and 1.066 Å for 8 CN. On the basis of SrNb₂O₆ doping with Eu³⁺, the Eu³⁺ ions will take place as an intermediate atom, while some Eu³⁺ ions will replace the ions of Sr²⁺. A small amount of Eu³⁺ ions lead to the formation of a minor amount of EuNbO₄ for 6 mol% sample, as seen in the XRD pattern in Fig. 1. Also, the increase in positive charge due to the inclusion of Eu³⁺ ions will affect the lattice charge balance in the lattice.

Fig. 2 Schematic illustration of SrNb₂O₆ crystal in the direction of the b axis.

SEM micrographs of MNb₂O₆:Eu³⁺ structures (M = Sr, Cd, Ni) for 3 mol% concentration of Eu³⁺ are presented in Fig. 3(a–d). SEM micrographs of Eu³⁺-doped SrNb₂O₆ are given in Fig. 3(a and b) at 2000× and 20 000× magnifications, respectively. The grain morphologies with rod-like shapes exhibited the characteristic structure of SrNb₂O₆.^7,8 The thicknesses and lengths of the rods for all the samples varied between 0.06–0.68 and 0.21–12 μm, respectively. The grain shape and size of CdNb₂O₆:Eu³⁺ and NiNb₂O₆Eu³⁺ differed in the SEM micrographs given in Fig. 3(c and d) and at 20 000× magnification, respectively. While the grains of Eu³⁺-doped CdNb₂O₆ were irregular, agglomerated, and plate-like shape, the grain sizes varied in the range of 0.5–4 μm. The grain morphologies of Eu³⁺-doped NiNb₂O₆ were round-like shape and relatively less agglomerated, with small grain sizes in the range of 0.1–1 μm. The EDS spectra and elemental compositions (at% and wt%) for 3 mol% Eu³⁺-doped MNb₂O₆ (M = Sr, Cd, Ni) samples are given in Fig. 4(a–c), respectively. The EDS atomic compositions (%) of the samples are highly compatible with the theoretical atomic values (%).

Fig. 3 SEM results of 3 mol% Eu³⁺-doped MNb₂O₆ samples: (a) SEM micrograph for SrNb₂O₆:Eu³⁺ at low (2000×) magnification, and SEM micrographs at high (20 000×) magnification: (b) SrNb₂O₆:Eu³⁺, (c) CdNb₂O₆:Eu³⁺, and (d) NiNb₂O₆:Eu³⁺.

Fig. 4 EDS elemental compositions (wt%, at%) and theoretical atomic values (%) for 3 mol% Eu³⁺-doped MNb₂O₆ samples: (a) SrNb₂O₆:Eu³⁺, (b) CdNb₂O₆:Eu³⁺, and (c) NiNb₂O₆:Eu³⁺.

3.2 Photoluminescence and radioluminescence analyses

Fig. 5(a–c) show the CT bands and PL excitations of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors recorded at emission wavelengths of 615.7, 616.2, and 614.0 nm, respectively. PL excitations of the phosphors were assigned to the transitions of ⁷F₀ → ⁵D₄, ⁷F₀ → ⁵G_j, ⁷F₀ → ⁵L₆, ⁷F₀ → ⁵D₃, and ⁷F₀ → ⁵D₂. The broad CT bands with peak centers at 277.4 nm (SrNb₂O₆:Eu³⁺), 276.6 nm (CdNb₂O₆:Eu³⁺), and 277.7 nm (NiNb₂O₆:Eu³⁺) could be attributed to the Eu³⁺–O²⁻ charge transfer, which results from the transition of 2p electrons of O²⁻ to the empty 4f orbitals of Eu³⁺ ion.^5,6 Fig. 6(a–c) show the PL emissions of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors monitored at the ⁵D₀ → ⁷F₀, ⁵D₀ → ⁷F₁, ⁵D₀ → ⁷F₂, ⁵D₀ → ⁷F₃, ⁵D₀ → ⁷F₄ transitions with the excitation of 393.6, 463.8, and 394.0 nm, respectively. The PL emissions of the CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ increased up to 6 mol%, while the SrNb₂O₆:Eu³⁺ emission decreased after 3 mol% due to the concentration quenching or nonradiative energy transfer, which originated from the host defects and the distance RE³⁺–RE³⁺.^5,6 In all the examples, the emission intensities of the ⁵D₀ → ⁷F₂ (electric dipole) transition were higher than the ⁵D₀ → ⁷F₁ (magnetic dipole) transition. This indicates that the Eu³⁺ ions occupied crystallographic sites without an inversion center.^19,20

Fig. 5 PL excitation spectra and charge-transfer bands (CTB) of the phosphors: (a) SrNb₂O₆:Eu³⁺, (b) CdNb₂O₆:Eu³⁺, and (c) NiNb₂O₆:Eu³⁺.

Fig. 6 PL emission spectra of the phosphors: (a) SrNb₂O₆:Eu³⁺, (b) CdNb₂O₆:Eu³⁺, and (c) NiNb₂O₆:Eu³⁺.

The radioluminescence (RL) emissions of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors were monitored with the ⁵D₀ → ⁷F₀, ⁷F₁, ⁷F₂, ⁷F₃, ⁷F₄ transitions under X-ray irradiation, as shown in Fig. 7(a–c). Also, the ⁵D₀ → ⁷F₅ and ⁵D₀ → ⁷F₆ transitions detected with RL are given in the inset figures in Fig. 7(a–c). The RL emissions of the phosphors showed a similarity as in the PL and decreased at 6 mol% of SrNb₂O₆:Eu³⁺. Additionally, in the RL emissions of SrNb₂O₆:Eu³⁺, an extra intensity increase or asymmetry (R) increase occurred at the ⁵D₀ → ⁷F₂ transition. Therefore both mechanisms could be compared based on the asymmetry, and the R-value variation of PL–RL emissions depending on the Eu³⁺ concentration is given Fig. 8. The R-values of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors were found to be 4.38–4.44, 6.58–7.13, 5.68–6.78 for PL, and 6.56–7.15, 6.38–7.42, 6.03–6.89 for RL, respectively. The R-value or (⁵D₀ → ⁷F₂)/(⁵D₀ → ⁷F₁) ratio can be used as an indicator to predict the symmetry of the local environment of Eu³⁺ ion; whereby a high R-value indicates an increase in Eu³⁺–O²⁻ covalency and a low symmetry/more distorted environment.⁵ As seen in Fig. 8, the R-values of the PL–RL mechanism for CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ were close to each other. However, the R-value of RL for SrNb₂O₆:Eu³⁺ was about 50% higher than the R-value of PL. The relatively low asymmetry of SrNb₂O₆:Eu³⁺ in PL increased in RL and reached levels close to other phosphors. Although the same basic transitions occur in PL and RL emissions, they do not contribute equally to the excitation mechanisms. X-ray irradiation of RL does not excite the CTB of the host but instead causes the formation of electron–hole pairs in its conduction and valence bands that recombine at the emission center, in which the energy is ultimately deposited.^7,21 X-ray irradiation can change the dipole moments and emission spectrum of Eu³⁺ in the host. Thus, although the symmetry of the Eu³⁺ local environment decreases to some extent with the high-energy photons of RL, the transformation will likely be temporary and the Eu³⁺ local environment symmetry will return to its initial state when X-ray irradiation is interrupted.^7,21

Fig. 7 RL emission spectra of the phosphors: (a) SrNb₂O₆:Eu³⁺, (b) CdNb₂O₆:Eu³⁺, and (c) NiNb₂O₆:Eu³⁺. The ⁵D₀ → ⁷F_j (j = 0, 1, 2, 3, 4) transitions between 570–730 nm, and (inset figures) the ⁵D₀ → ⁷F_j (j = 5, 6) transitions between 690–850 nm.

Fig. 8 R-Value variation for PL and RL mechanisms depending on the Eu³⁺ concentration.

The possible PL and RL mechanisms of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors are schematically illustrated in Fig. 9(a and b), respectively. Fig. 9(a) shows the PL mechanism and indicates the possible pathways in the PL process. Excitation may consist of two processes, either by transferring the valence band electrons to the charge-transfer band (CTB) or by jumping to the higher levels of Eu³⁺.⁶ Excitation with charge transfer takes place at 276–278 nm while direct excitation of Eu³⁺ occurs at 393.6, 463.8, 394.0 nm. After excitation, the excited electrons come to the lowest excited level (⁵D₀) of Eu³⁺ by a nonradiative transition. Then it can make (⁵D₀ → ⁷F_J) transitions to the lower levels, in which emissions of Eu³⁺ would occur. The RL possible mechanism of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors is presented in Fig. 9(b). The electrons of Eu³⁺ in the ground state (⁷F_J) promote the conduction band (CB), while the holes stay in the valence band (VB). X-ray induced electrons move in the CB and then part of them transfer to the excitation levels of Eu³⁺. Also, the formation of traps is possible in the RL mechanism, and a part of the electrons can be captured by the traps, although this is not shown in the figure. Finally, when X-ray-induced electrons move to the ground state, they would be released at the appropriate energy level in which RL emissions take place with the recombination of the electrons and holes.

Fig. 9 Schematic energy level diagram to illustrate the possible (a) PL and (b) RL mechanisms.

3.3. Judd–Ofelt analysis and radiative properties

Judd–Ofelt (JO) intensity parameters are critical to evaluate the performance of luminescent materials.^22,23 Among the RE³⁺ ions, Eu³⁺ has a unique feature for calculation of the JO intensity parameters due to the magnetic dipole transition (⁵D₀ → ⁷F₁) which is independent of the environment and can be used as a reference for transitions arising from the ⁵D₀ level. Using the emission spectra for Eu³⁺-doped materials, the JO intensity parameters Ω_J (J = 2, 4, 6) can be calculated by eqn (1):²⁴where V₁, V_J and I₁, I_J are the frequencies and integrated intensities of the ⁷F₁ and ⁷F_J (J = 2, 4, 6) transitions respectively, e = 4.803 × 10⁻¹⁰ esu (1 esu = N^−5/2 cm = gr^1/2 cm^3/2 s⁻¹) is the elementary charge, n is the refractive index, S_MD = 9.6 × 10⁻⁴² esu² cm² is the line strength of the magnetic dipole, which was taken from ref. 24, and |〈J‖U^J‖J〉|² represent the double-reduced matrix elements of the unit tensor operators, which are independent of the local environment of the ion. All the reduced matrix elements (|〈J‖U^J‖J′〉|²) for electric-dipole (ED) transitions arising from the ⁵D₀ level are zero, except for the levels ⁷F_J, (where J = 2, 4, 6) U² = 0.0032, U⁴ = 0.0023, and U⁶ = 0.0002.²⁴ The ⁵D₀ → ⁷F_J (J = 1, 2, 4, 6) transitions are used in the calculation of the JO parameters for the radiative transition probability. However, the ⁵D₀ → ⁷F_J (J = 0, 3, 5) transitions are forbidden and are not included in the determination of the transition probabilities.²⁴ According to Judd–Ofelt theory, the spontaneous transition probability (A) is related to its magnetic dipole strength (S_MD) and electric-dipole strength (S_ED), which can be calculated by eqn (2):^22,23where J is the total angular momentum of the initial state, and h is Planck's constant. The electric-dipole line strengths (S_ED) can be calculated from the JO parameters by eqn (3):⁷where |〈J‖U^J‖J〉|² is the symbol of the squared reduced matrix elements, and χ_ED and χ_MD are the local field corrections for the ED and MD transitions, which can be found by eqn (4) and (5), respectively:

χ_MD = n³ (5)

where n is the refractive index. The n = 1.854 value for SrNb₂O₆ was taken from ref. 7. The n value for CdNb₂O₆ and NiNb₂O₆ can be estimated from the Lorenz–Lorentz formula eqn (6):^25–29where M is the molar mass of the compound, l_i is the atomic number of the element in the nominal chemical formula of the compound, ρ is the density, and r_i is the specific refraction of the compound. The refraction index (n) values for CdNb₂O₆ and NiNb₂O₆ were determined as 1.958 and 2.003, respectively.

The spectral results of the phosphors, JO intensity parameters (Ω₂, Ω₄), radiative transition probabilities (A(J,J′)), and total radiative transition probabilities ( or A_r) are given in Table 1. The Ω₂ parameter shows the short-range effects, while the Ω₄ parameter shows the long-range effects.^25,31 The effects of the Ω₂ parameter are short-range and are associated with the covalence and environmental changes of the Eu³⁺ ion symmetry.^19,20 The high value of Ω₂ or the trend Ω₂ > Ω₄ is evaluated to show the high covalence of the Eu³⁺–O²⁻ bond character (ionic or covalent) and the high asymmetry or distortion of the local symmetry of Eu³⁺ sites.^7,31 Accordingly, a high Ω₂ parameter or Ω₂ > Ω₄ tendency for all the phosphors can be attributed to the low symmetry Eu³⁺ ion and the more covalent or less ionic character of the Eu³⁺–O²⁻ bond. On the other hand, the effects of the Ω₄ parameter are long-range and are not directly related to the symmetry of the Eu³⁺ ion, but the electron density on the surrounding ligands.^30,31 In all phosphors, the parameter of Ω₄ decreased with increasing Eu³⁺ concentration, which indicated an increase in electron density in the ligands.

Table 1 J–O parameters (Ω₂, Ω₄), radiative transition and total transition probabilities (A(J,J′), (A_r)), branching ratios ((β_cal), (β_exp)), and branching ratio differences (%) of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors

Phosphor	Eu^3+ conc. (mol%)	Eu^3+ transitions	Ω2 (10^−20 cm^2)	Ω4 (10^−20 cm^2)	A(J,J′) (s^−1)	Ar (s^−1)	βcal (%)	βexp (%)	Difference (%)
SrNb2O6:Eu^3+	0.5	^5D0 → ^7F1	7.148	5.837	92.101	694.935	13.25	14.13	6.20
		^5D0 → ^7F2			429.055		61.74	63.24	2.37
		^5D0 → ^7F4			173.779		25.01	22.63	9.49
	3	^5D0 → ^7F1	7.209	5.818	92.026	697.946	13.19	14.05	6.16
		^5D0 → ^7F2			432.719		61.99	63.49	2.35
		^5D0 → ^7F4			173.201		24.82	22.46	9.50
	6	^5D0 → ^7F1	7.172	5.779	92.194	694.717	13.27	14.15	6.80
		^5D0 → ^7F2			430.474		61.96	63.44	2.33
		^5D0 → ^7F4			172.049		24.77	22.41	9.52
CdNb2O6:Eu^3+	0.5	^5D0 → ^7F1	7.360	1.513	107.954	699.085	15.44	16.09	4.04
		^5D0 → ^7F2			538.924		77.09	77.40	0.39
		^5D0 → ^7F4			52.207		7.47	6.51	12.78
	3	^5D0 → ^7F1	7.788	1.396	107.933	726.332	14.86	15.47	3.95
		^5D0 → ^7F2			570.238		78.51	78.75	0.31
		^5D0 → ^7F4			48.161		6.63	5.78	12.86
	6	^5D0 → ^7F1	7.852	1.388	107.916	730.720	14.77	15.37	3.94
		^5D0 → ^7F2			574.902		78.67	78.92	0.30
		^5D0 → ^7F4			47.902		6.56	5.71	12.86
NiNb2O6:Eu^3+	0.5	^5D0 → ^7F1	8.529	3.235	116.395	926.132	12.57	13.19	4.75
		^5D0 → ^7F2			685.886		74.06	75.00	1.25
		^5D0 → ^7F4			123.851		13.37	11.81	11.72
	3	^5D0 → ^7F1	9.658	2.270	116.375	979.958	11.88	12.40	4.21
		^5D0 → ^7F2			776.661		79.25	79.82	0.70
		^5D0 → ^7F4			86.922		8.87	7.79	12.21
	6	^5D0 → ^7F1	10.023	2.726	116.340	1043.388	11.15	11.66	4.37
		^5D0 → ^7F2			822.662		78.85	79.54	0.88
		^5D0 → ^7F4			104.386		10.00	8.80	12.06

---

The branching ratio (β or β_cal) of the Judd–Ofelt theory can be found from the radiative transition probability (A(J,J′)) and total radiative transition probability by eqn (7):⁷

The branching ratios (β_cal) of ⁵D₀ → ⁷F₂ transition for CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ were found as 77.09–78.67% and 74.06–79.25%, respectively. The SrNb₂O₆:Eu³⁺ had a relatively low branching ratio ranging from 61.74–61.99%, for which the results are given in Table 1. The β_cal values of the ⁵D₀ → ⁷F₂ transition for all phosphors were over 60%. The high value of branching ratio (β > 50) indicated a potential laser emission.^7,24 The branching ratios (β_cal) of the Judd–Ofelt theory can be evaluated using experimental branching ratios (β_exp), which can be obtained directly from the emission spectrum. On the other hand, the intrinsic error of the Judd–Ofelt theory in describing spectral intensities is usually about 15%.^7,32 Accordingly, differences between the experimental (β_exp) and calculated values (β_cal) for all samples occurred between 0.3–12.86%, and they were all below 15% (Table 1).

In the Judd–Ofelt calculation, the Ω₂ and Ω₄ parameters were used, but the Ω₆ parameter was not included since the ⁵D₀ → ⁷F₆ transition remaining in the infrared region could not be detected by PL. Similar studies have been reported using the Ω₂ and Ω₄ parameters for JO calculations.^24,30 The ⁵D₀ → ⁷F₆ transition outside the visible spectrum is very weak in many materials.²⁴ In this study, the weak intensity of the ⁵D₀ → ⁷F₆ transition detected by RL in Fig. 7(a–c) supports this situation. Based on the ⁵D₀ → ⁷F₆ transition observed by RL, the Ω₆ parameters were determined for the samples with the highest emissions intensity. The Ω₆ parameters and total radiative transition probabilities ( or A_r) of the ⁵D₀ → ⁷F₆ transition were estimated as 2.226 × 10⁻²⁰, 1.103 × 10⁻²⁰, 0.155 × 10⁻²⁰ cm², and 3.636, 2.116, 0.341 s⁻¹ for 3 mol% (SrNb₂O₆:Eu³⁺), 6 mol% (CdNb₂O₆:Eu³⁺), 6 mol% (NiNb₂O₆:Eu³⁺), respectively. The branching ratios (β_cal) of the ⁵D₀ → ⁷F₆ transition (or the ratio of the ⁵D₀ → ⁷F₆ transition probability to the total radiative transition probability) were found in the range of 0.03–0.52%. Also, the branching ratios (β_exp) of the ⁵D₀ → ⁷F₆ transition from the emission spectra were between 0.18–0.40%. Accordingly, based on the determination of the branching ratios using the RL spectra, the effect of the ⁵D₀ → ⁷F₆ transition on the quantum efficiency or other radiative parameters was negligibly minimal, and its exclusion will not have a significant effect.

Fig. 10(a–c) show the decay curves of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors for ⁵D₀ → ⁷F₂ transitions (λ_em = 615.7, 616.2, 614.0 nm) recorded with excitations at 393.6, 463.8, and 394.0 nm, respectively. The observed lifetime or average decay (τ) curves can be established by the following two-time temporal dependence in eqn (8):⁶

where the luminescence intensities are I^L_(t) and I^L₍₀₎, the fitting constants are B₁ and B₂, and the short and long lifetimes for the exponential components are τ₁ and τ₂ respectively. The average (observed) lifetime (τ) can be calculated through eqn (9) according to the variables in eqn (8):⁶

Fig. 10 PL lifetime curves for ⁵D₀ → ⁷F₂ transition of the phosphors: (a) SrNb₂O₆:Eu³⁺, (b) CdNb₂O₆:Eu³⁺, and (c) NiNb₂O₆:Eu³⁺.

The observed lifetimes (τ) of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors for the ⁵D₀ level varied from 914.36, 673.35, 119.08 to 828.69, 573.29, 113.75 μm, respectively (Table 2). The observed lifetime (τ) represents the sum of nonradiative (τ_nr) and the radiative (τ_r) lifetimes, which are expressed by eqn (10). The value of the radiative (experimental) lifetime (τ_r) can be determined from the total radiative transition probability (A_r) by eqn (11):

τ = τ_r + τ_nr (10)

Table 2 The radiative lifetimes (τ_r), theoretical radiative lifetimes (τ_th), observed lifetimes (τ), quantum efficiencies (η_QE), theoretical quantum efficiencies (η^th_QE), and quantum efficiency differences (%) of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors

Phosphor	Eu^3+ conc. (mol%)	τr (μs)	τth (μs)	τ (μs)	ηQE (%)	η^thQE (%)	Difference (%)
SrNb2O6:Eu^3+	0.5	1438.98	1429.07	914.36	63.54	63.98	0.69
	3	1432.78	1426.16	886.52	61.87	62.16	0.46
	6	1439.43	1433.49	828.69	57.57	57.81	0.41
CdNb2O6:Eu^3+	0.5	1430.44	1344.79	673.35	47.07	50.07	5.99
	3	1376.78	1308.06	609.02	44.24	46.56	4.99
	6	1368.51	1302.47	573.29	41.89	44.02	4.83
NiNb2O6:Eu^3+	0.5	1080.76	1002.41	119.08	11.03	11.88	7.16
	3	1020.45	989.21	115.91	11.36	11.72	3.06
	6	958.42	930.47	113.75	11.87	12.22	2.92

---

The radiative lifetime can also be found theoretically by direct calculation from eqn (12), which is related to the emission spectrum of Eu³⁺.^24,32

where τ_th is the theoretical radiative lifetime, I₁ is the integrated intensity of the ⁵D₀ → ⁷F₁ transition, I_tot is the total integrated intensity, and n is the refractive index. The quantum efficiency (η_QE) of a phosphor can be defined by the ratio of the number of photons emitted (observed lifetime: τ) to the number of photons absorbed (radiative lifetimes: τ_r). Accordingly, the quantum efficiency (η_QE) and theoretical quantum efficiency (η^th_QE) can be determined by eqn (13) and (14), respectively:

A comparison of the radiative lifetimes (τ_r, τ_th) and quantum efficiencies (η_QE, η^th_QE) of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors was performed and the results are given in Table 2. The η_QE% values for the SnNb₂O₆:Eu³⁺ and CdNb₂O₆:Eu³⁺ were estimated as follows: 57.57–63.54% and 41.89–47.07%, respectively. However, the η_QE% value of NiNb₂O₆:Eu³⁺ showed a low efficiency and varied between 11.22–11.88%. The differences of the η_QE% (calculated-JO) and η^th_QE (theoretical) for SnNb₂O₆:Eu³⁺ varied between 0.41–0.69%, whereas the compatibility of the efficiencies for CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺ decreased somewhat and the difference varied between 4.83–5.99 and 2.92–7.16, respectively (Table 2).

The stimulated emission cross-section (σ_e) is an important factor to predict the laser performance of a material. The stimulated emission cross-section (σ_e) between the states J and J′ having a probability of A(J,J′) is given by eqn (15):^7,20

where c is the velocity of light, λ_p is the emission peak wavelength, Δλ_eff is the effective bandwidth of the emission transition and n is the refractive index of the matrix. The stimulated emission cross-section (σ_e), gain bandwidth (σ_e × Δλ_eff), the effective bandwidth of emission transition (Δλ_eff), and the optical gain (σ_e × τ) results are given in Table 3. The large σ_e value for the potential laser performance at room temperature is an attractive feature for low threshold and high gain laser applications.⁷ The σ_e parameters of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors were found as ⁷F₂ > ⁷F₄ > ⁷F₁. The product of the σ_e and Δλ_eff is another important parameter and helps predict the gain bandwidth (σ_e × Δλ_eff) of the optical amplifier. The high values of the product are better for the performance of the amplifier.⁷ The gain bandwidths (σ_e × Δλ_eff) of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors with the highest emission were determined as follows: 24.182 × 10⁻²⁸, 28.674 × 10⁻²⁸, and 38.647 × 10⁻²⁸ cm³, respectively, corresponding to ⁵D₀ → ⁷F₂ transition. The product of the σ_e and observed lifetime (τ) is an important consideration for high optical amplifier gain.^7,33 The optical gains (σ_e × τ) of MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) varied between 19.283–21.750 × 10⁻²⁵, 13.790–15.859 × 10⁻²⁵, and 3.518–3.987 × 10⁻²⁵ cm² s, respectively. The large values of the stimulated emission cross-section, gain bandwidth, optical gain, and lifetime support that MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors can be helpful for the improvement of the red lasing and red color applications for display devices.

Table 3 The effective bandwidth of the emission transitions (Δλ_eff), stimulated emission cross-sections (σ_e), gain bandwidths (σ_e × Δλ_eff), and optical gains (σ_e × τ) of the MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors

Phosphor	Eu^3+ conc. (mol%)	Eu^3+ transitions	Δλeff (nm)	σe (×10^−22 cm^2)	σe × Δλeff (×10^−28 cm^3)	σe × τ (×10^−25 cm^2 s)
SrNb2O6:Eu^3+	0.5	^5D0 → ^7F1	12.10	3.627	4.388
		^5D0 → ^7F2	10.08	23.788	23.977	21.750
		^5D0 → ^7F4	11.03	14.452	15.933
	3	^5D0 → ^7F1	12.15	3.613	4.389
		^5D0 → ^7F2	10.49	23.057	24.182	20.441
		^5D0 → ^7F4	11.21	14.161	15.880
	6	^5D0 → ^7F1	12.12	3.618	4.387
		^5D0 → ^7F2	10.34	23.269	24.056	19.283
		^5D0 → ^7F4	11.13	14.168	15.775
CdNb2O6:Eu^3+	0.5	^5D0 → ^7F1	14.36	3.233	4.641
		^5D0 → ^7F2	11.41	23.552	26.880	15.859
		^5D0 → ^7F4	12.42	3.683	4.574
	3	^5D0 → ^7F1	15.19	3.055	4.642
		^5D0 → ^7F2	11.50	24.727	28.442	15.059
		^5D0 → ^7F4	12.80	3.297	4.219
	6	^5D0 → ^7F1	15.85	2.929	4.643
		^5D0 → ^7F2	11.92	24.055	28.674	13.790
		^5D0 → ^7F4	12.91	3.251	4.197
NiNb2O6:Eu^3+	0.5	^5D0 → ^7F1	12.96	3.657	4.737
		^5D0 → ^7F2	10.91	29.546	32.222	3.518
		^5D0 → ^7F4	9.11	11.060	10.081
	3	^5D0 → ^7F1	13.39	3.537	4.737
		^5D0 → ^7F2	10.90	33.487	36.486	3.881
		^5D0 → ^7F4	9.63	7.345	7.075
	6	^5D0 → ^7F1	13.49	3.512	4.738
		^5D0 → ^7F2	11.03	35.054	38.647	3.987
		^5D0 → ^7F4	10.93	7.777	8.497

---

4. Conclusion

MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) crystal powders synthesized by the molten salt method have an orthorhombic/monoclinic symmetry. SEM micrographs of Eu³⁺-doped SrNb₂O₆ exhibited characteristic nanorods for the grain structure, while the grain morphologies of Eu³⁺-doped CdNb₂O₆ and NiNb₂O₆ differed in size and shape. The emissions of the phosphors were monitored in PL and RL with the ⁵F₀ → ⁷F_J (J = 0–4) and ⁵F₀ → ⁷F_J (J = 0–6) transitions of Eu³⁺, respectively. The RL emissions of SrNb₂O₆:Eu³⁺ exhibited an extra increase at the ⁵D₀ → ⁷F₂ (electric dipole) transition, which was related to the decrease in local symmetry of Eu³⁺; however, this same effect was not seen for CdNb₂O₆:Eu³⁺ and NiNb₂O₆:Eu³⁺. By determining the Judd–Ofelt intensity parameters (Ω₂, Ω₄) from the emission spectrum, we calculated the radiative transition probabilities (A(J,J′)), branching ratios (β_cal), and radiative lifetimes (τ_r). The branching ratio difference of the measured (β_exp) from the emission spectrum and that calculated (β_cal) from the JO parameters for all samples varied by 0.3–12.86%, which were all less than 15%. The high Ω₂ value or Ω₂ > Ω₄ tendency for all the phosphors indicated the low symmetry Eu³⁺ ion and the more covalent character of the Eu³⁺–O²⁻ bond. The order of quantum efficiencies (η_QE%) was found to be: SrNb₂O₆:Eu³⁺ > CdNb₂O₆:Eu³⁺ > NiNb₂O₆:Eu³⁺. The quantum efficiencies (η_QE) of the phosphors deviated slightly from the theoretically calculated quantum efficiencies (η^th_QE). The bandwidths (σ_e × Δλ_eff) and optical gains (σ_e × τ) were found to follow the trend: SrNb₂O₆:Eu³⁺ < CdNb₂O₆:Eu³⁺ < NiNb₂O₆:Eu³⁺ and SrNb₂O₆:Eu³⁺ > CdNb₂O₆:Eu³⁺ > NiNb₂O₆:Eu³⁺, respectively. These results point out that MNb₂O₆:Eu³⁺ (M = Sr, Cd, Ni) phosphors with strong emissions at ⁵D₀ → ⁷F₂ transition could be interesting candidates for RE-doped red phosphor applications and for future study.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We would like to thank to Marmara University Research Fund (Project No: FEN-A–110718–0393) for the financial support for this study.

References

1. S. Saha, S. Das, U. K. Ghorai, N. Mazumder, B. K. Gupta and K. K. Chattopadhyay, Dalton Trans., 2013, 42, 12965–12974.
2. C. Yu, B. Chen, X. Zhang, X. Li, J. Zhang, S. Xu, H. Yu, J. Sun, Y. Cao and H. Xia, Phys. Chem. Chem. Phys., 2020, 22, 7844–7852.
3. Y. Zhang, B. Chen, S. Xu, X. Li, J. Zhang, J. Sun, X. Zhang, H. Xia and R. Hua, Phys. Chem. Chem. Phys., 2018, 20, 15876–15883.
4. B. Tian, B. Chen, Y. Tian, X. Li, J. Zhang, J. Sun, H. Zhong, L. Cheng, S. Fu, H. Zhong, Y. Wang, X. Zhang, H. Xiac and R. Hua, J. Mater. Chem. C, 2013, 1, 2338–2344.
5. M. İlhan, Synthesis, Int. J. Appl. Ceram. Technol., 2017, 14, 1134–1143.
6. M. İlhan and İ. Ç. Keskin, Dalton Trans., 2018, 47, 13939–13948.
7. M. İlhan and İ. Ç. Keskin, Phys. Chem. Chem. Phys., 2020, 22, 19769–19778.
8. M. K. Ekmekçi, M. İlhan, A. Ege and M. Ayvacıklı, J. Fluoresc., 2017, 27, 973–979.
9. J. Xue, Y. Guo, B. K. Moon, S. H. Park, J. H. Jeong, J. H. Kim and L. Wang, Opt. Mater., 2017, 66, 220–229.
10. R. C. Ropp, Encyclopedia of the Alkaline Earth Compounds, Elsevier Publications, Oxford, UK, 2013.
11. I. S. Cho, S. Lee, J. H. Noh, D. W. Kim, D. K. Lee, H. S. Jung, D. W. Kim and K. S. Hong, J. Mater. Chem., 2010, 20, 3979–3983.
12. K. N. Singh and P. K. Bajpai, World J. Condens. Matter Phys., 2011, 1, 37–48.
13. M. K. Ekmekçi, M. Erdem, M. İlhan, A. S. Başak and A. Mergen, Optik, 2016, 127, 1918–1921.
14. M. K. Ekmekçi, M. Erdem, A. S. Başak, M. İlhan and A. Mergen, Ceram. Int., 2015, 127, 9680–9685.
15. M. G. B. Drew, R. J. Hobson and V. T. Padayatchy, J. Mater. Chem., 1993, 3, 889–892.
16. M. Sato and Y. J. Hama, J. Solid State Chem., 1995, 118, 193–198.
17. I. S. Cho, S. Lee, J. H. Noh, D. W. Kim, H. S. Jung, D. W. Kim and K. S. Hong, Cryst. Growth Des., 2010, 10, 2447–2450.
18. K. L. Keester, R. R. Neurgaonkar, T. C. Lim and E. J. Staples, Mater. Res. Bull., 1980, 15, 821–826.
19. R. G. A. Kumar, S. Hata, K. Ikeda and K. G. Gopchandran, RSC Adv., 2016, 6, 67295–67307.
20. R. Saraf, C. Shivakumara, S. Behera, H. Nagabhushana and N. Dhananjaya, RSC Adv., 2015, 5, 4109–4120.
21. K. Wahid, M. Pokhrel and Y. Mao, J. Solid State Chem., 2017, 245, 89–97.
22. B. R. Judd, Phys. Rev., 1962, 127, 750.
23. G. S. Ofelt, J. Chem. Phys., 1962, 37, 511.
24. A. Ćirić, S. Stojadinović, M. Sekulić and M. D. Dramićanin, J. Lumin., 2019, 205, 351–356.
25. M. Born and E. Wolf, Principles of Optics, Pergamon Press, Oxford, London, Edinburg, New York, Paris, Frankfurt, 1968.
26. S. S. Batsanov, Structurnaya refractometria, Visshaya Shkola, Moscow, 1976.
27. R. D. Shannon and R. X. Fischer, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 73, 235111.
28. G. B. Bokii and M. A. P. Koshits, X-ray Structure Analysis, MSU, Moscow, 1964.
29. A. S. Korotkov and V. V. Atuchin, Opt. Commun., 2008, 281, 2132.
30. M. İlhan and İ. Ç. Keskin, Physica B, 2020, 85, 412106.
31. S. Dutta, S. Som and S. K. Sharma, RSC Adv., 2018, 8, 53–67.
32. M. H. V. Werts, R. T. F. Jukes and J. W. Verhoeven, Phys. Chem. Chem. Phys., 2002, 4, 1542–1548.
33. R. J. M. Gopinath, S. Gopi, S. M. Simon, A. C. Saritha, P. R. Biju, C. Joseph and N. V. Unnikrishnan, RSC Adv., 2020, 10, 20057–20066.

---