Eu³⁺ local site analysis and emission characteristics of novel Nd₂Zr₂O₇:Eu phosphor: insight into the effect of europium concentration on its photoluminescence properties

Abstract

A new zirconate-based pyrochlore-type red phosphor, Nd₂Zr₂O₇:Eu³⁺, was synthesized using the gel-combustion method. The un-doped sample showed visible emission due to the presence of oxygen vacancies. On doping with europium ions, it was found that Eu was distributed into both Nd³⁺ as well as Zr⁴⁺ sites based on emission and lifetime spectroscopy with a lifetime value of 3.35 and 1.14 ms, respectively, although the population at the Zr⁴⁺ site is more than that at the Nd³⁺ site. This is also justified by the presence of two stark components in the ⁵D₀–⁷F₀ transition. The presence of host emission in the spectrum of Nd₂Zr₂O₇:Eu³⁺ indicated incomplete energy transfer process at low levels of europium doping. Moreover, the efficiency of energy transfer increased as the doping percentage increased, and complete host-dopant energy transfer took place at 5.0% Eu³⁺ doping. Based on time resolved emission spectroscopy (TRES), the emission spectrum of europium into the Nd³⁺ sites as well as at Zr⁴⁺ sites was elucidated. Calculation of the Judd–Ofelt parameter illustrated that at all europium concentration, the local surrounding area lacks inversion symmetry (asymmetric environment) and the exact point group symmetry was found to be C₄. The red purity and high quantum efficiency of Nd₂Zr₂O₇:Eu³⁺ projects it to be a new materials for commercial application in white light emitting diodes.

---

1. Introduction

Recently lanthanide ion doped inorganic materials have become the focal point of materials research because of their unique optical, electronic and magnetic properties arising from sharp and narrow f–f transitions of the 4f orbitals in Ln³⁺. Some of the very recent applications include phosphors,¹ metastable phase stabilizers,² electrocatalysts,³ upconversion,⁴ sensitive molecular sensors,⁵ medical applications,⁶ biological assays based on mass cytometry,⁷ etc. Among the lanthanide doped inorganic materials, trivalent europium-based luminescence has gained significant importance because of its simple electronic energy level scheme and the hypersensitive nature of the electronic dipole transition (⁵D₀ → ⁷F₂) that can be used as a spectroscopic probe for host-dopant energy transfer dynamics,⁸ site swapping,⁹ point group symmetry,¹⁰ covalency/polarizability effect of ligands,¹¹ intricate structure-photoluminescence correlation,¹² local site occupancy,¹³ etc.

Hosts play an important role for highly efficient luminescence materials in terms of color tunability, high lumen output, low power usage, etc. In our laboratory, based on usage of different kinds of luminescence, hosts have been explored such as AB₂O₄ types spinels,¹⁴ hexagonal perovskite,¹⁵ orthorhombic perovskite/double perovskite,^1,9 molybdate/tungstate,^3,8,16 and silicate/phosphate.^10,17

A₂B₂O₇ types of pyrochlore oxides with lots of favorable properties such as high chemical and thermal stability, lattice stiffness and the ability to accommodate various dopant ions can be a very good candidate for a luminescent host.

It can exist in two different crystallographic structures (a) ideal pyrochlore and (b) distorted fluorite structure. These structures are inter-convertible based on applied temperature and pressure or by chemical doping.

In the fluorite kind of structure, A and B ions are oriented in a disordered array across the 4a sites whereas oxygen and vacancies are oriented in the same manner but across the 8c sites. On the other hand, in the case of the ideal pyrochlore structure (S.G. Fm3̄m), A and B ions and oxygen atoms are orderly arranged.

The stability of the particular phase for a given composition (pyrochlore and fluorite structure) is governed by radius ratio rules (r_A/r_B) with the ideal pyrochlore being stable in the region r_A/r_B = 1.46–1.78.¹⁸ Nd₂Zr₂O₇ (with r_A/r_B ∼1.54) has been known to adopt the pyrochlore structure when processed using high temperature solid state diffusion, but it is known to undergo a thermally induced pyrochlore to fluorite (order–disorder) transition at 2300 °C.¹⁹

In the last decade or so, Nd₂Zr₂O₇ turns out to be an important technological material not only from an application point of view, but also from a research perspective because of its various suitable and advantageous properties. These properties includes a high coefficient of thermal expansion (CTE), low thermal conductivity, metal to insulator transitions, geometrically frustrated Kagome-type spin lattices, dielectric properties,²⁰ etc. These properties have led to applications in various diversified areas such as magnetism,²¹ solid oxide fuel cells,²² nuclear waste hosts,²³ photocatalysts,²⁴ solid state laser materials,²⁵ etc.

In order to synthesize a highly efficient phosphor material, the usual way is to incorporate lanthanide ion in a low phonon inorganic host. The photoluminescence is purely based on f–f or f–d transitions of the lanthanide ion. Different lanthanide ions, depending upon their energy level, emit in different regions of the color spectrum. Eu³⁺ is known to emit in the red region, Tb³⁺ in the green and Dy³⁺ can emit in the white region.²⁶ It is well-known that Eu³⁺ has very sharp and weak excitation peaks in the range of 300 to 450 nm due to the forbidden nature of the f–f transition; therefore, most of the reported Eu³⁺-doped phosphors are excited at narrow excitation wavelengths resulting in weak excitation of Eu-doped luminescent materials. Furthermore, the full width at half-maximum (FWHM) of the lanthanide absorption peaks is too sharp/narrow to withstand the shift in emission wavelength position of light emitting diode chips. As a result of this, white LEDs made from Eu³⁺-doped phosphors are poor.

In some cases, the host itself gives self-assisted emission that can be because of defect-related^3,14 or intrinsic charge transfer⁸ and they can effectively transfer their excitation energy to the dopant ion when certain lanthanide ions are incorporated into the material. In fact, host sensitized luminescence via host-dopant energy transfer has become an effective route to enhance the emission intensity of Eu³⁺ ion in a phosphor material.

Also, in a matrix where there are multiple sites for occupancies like in Nd₂Zr₂O₇, it becomes an imperative task to know how the dopant ion is going to distribute itself upon incorporation. Information on local site occupancy is very important to optimize the optical properties of a material, particularly in the case of europium because it yields site sensitive luminescence.^9,17,27,28

In Nd₂Zr₂O₇, Nd³⁺ has 8-fold coordination whereas Zr⁴⁺ has 6-fold coordination and the site symmetry around Nd³⁺ is D_3d. We have taken the Eu³⁺ ion as a spectroscopic probe for quite obvious reasons as explained in earlier in the introduction. Analyzing the information on the site distribution of the Eu³⁺ ion in the Nd₂Zr₂O₇ host involves determining whether it occupies a single site (Nd³⁺ or Zr⁴⁺) or multiple sites (both Nd³⁺ and Zr⁴⁺). Getting such information only by experimental observations can sometimes become very challenging due to the low level of dopant concentration and the localized nature of the structural distortions that arise due to the size and charge mismatch between dopant and lattice ion. Such a problem can be solved using theoretical simulations. Identifying the local structure around Eu³⁺ in Nd₂Zr₂O₇ and applying proper slicing to get the individual time resolved emission spectrum of Eu at Nd and Zr sites is also very important for color tunable LEDs. Although the criteria for size matching guarantees Eu³⁺ to occupy Nd³⁺, it is possible that some of the Eu³⁺ may also reside at Zr⁴⁺ sites. Also, this particular phosphor is prepared using a gel combustion route where one can control the size and shape by varying the oxidant to fuel ratio. Other techniques such as sol–gel, solid state and co-precipitation routes are not as advantageous as the combustion techniques. As far as the solid state reaction is concerned, the high impurity and poor powder characteristics represented by a coarse particle size, wide particle size distribution, irregular particle morphology and a high degree of inhomogeneity made this process unsuitable.²⁹

The sol–gel process requires a metal alkoxide as a precursor that is very expensive and extremely sensitive to heat, light and water, which makes it necessary to conduct the sol–gel experiment in dry boxes or clean rooms. Also, this method needs long heat-treatment times.³⁰ On the other hand, even simple methods like co-precipitation is very time consuming as it requires repeated washing to remove unwanted anions. So every method has its own advantage and disadvantage. Even in the case of the gel-combustion method, one has to very cautious while conducting experiments and no one can do such experiments without lab safety masks because combustion involves evolution of toxic gases like CO and NOx.

As such, there have been no reports pertaining to the photophysical properties of Nd₂Zr₂O₇ doped with varied concentrations of europium ion. This is, indeed, a new addition to the phosphor library wherein we have investigated the effect of europium ion concentration on host-dopant energy transfer process using time resolved photoluminescence. We have also used experimental measurements to decipher information on the local site occupied by the europium ion in the Nd₂Zr₂O₇ host. Other photophysical parameters such as site symmetry, radiative lifetime, quantum efficiency, Judd–Ofelt parameter and branching ratio were also evaluated.

2. Experimental

2.1. Synthesis: gel-combustion method

Un-doped and europium-doped (0.1 to 5.0 mol%) Nd₂Zr₂O₇ was prepared using a soft chemical sol–gel-combustion process. Nd₂O₃/Eu₂O₃ (high purity, SPEX industries inc. New Jersey, U.S.A) zirconium(IV) oxynitrate hydrate [ZrO(NO₃)₂·xH₂O, Alfa Aesar] and citric acid monohydrate [C₆H₈O₇·H₂O, Sigma Aldrich] were used as the starting materials. At first, both Nd₂O₃ and Eu₂O₃ oxide were converted into the nitrated form using 8 M HNO₃. Citric acid (CA) being a reaction fuel helps in the generation of a large amount of heat for combustion that drives the reaction between Nd³⁺ and Zr⁴⁺. Oxygen to fuel ratio (O/F) was fixed to 1 : 1.87 (fuel excess) based on the concept of propellant chemistry. Fuel being slightly in excess, an appropriate amount of NH₄NO₃ was added as an external oxidant.

First, all the three precursor components (Nd, Zr and CA) were dissolved in Millipore water. Then, they were mixed together and kept on a magnetic stirrer for about 8–10 h. This resulted in a completely homogenous solution. The gelation started upon raising the temperature to 80 °C. This gel was dried under shallow infrared light that resulted in the complete removal of water and leads to the formation of a highly condensed porous network. Then, the beaker was transferred into a muffle furnace and heated to 350 °C for 30 minutes. This resulted in the formation of a highly voluminous carbon black solid mass that was grounded into a fine powder with a mortar and pestle and placed in alumina crucibles to be heat treated to 800 °C for 6–7 h. For the Eu³⁺-doped sample, an appropriate amount of Eu₂O₃ was added at the initial stage.

2.2. Instrumentation

X-ray diffraction (XRD) patterns of the powdered europium-doped Nd₂Zr₂O₇ (NEZO) samples were recorded using a RIGAKU Miniflex-600 diffractometer operating in the Bragg–Brentano focusing geometry. Cu-Kα radiation source (λ = 1.5406 Å) was used as the X-ray source. The operating voltage and current of the instrument was kept at 40 kV and 30 mA, respectively. The XRD patterns were collected with a scan rate of 1°/minute. PL data were recorded on an Edinburgh CD-920 unit equipped with M 300 monochromators. Data acquisition and analysis were done by the F-900 software provided by Edinburgh Analytical Instruments, UK. A xenon flash lamp with frequency range of 10–100 Hz was used as the excitation source. Emission spectra for a particular sample were recorded with a lamp frequency of 100 Hz. Multiple scans (at least five) were taken to minimize the fluctuations in peak intensity and maximize the S/N ratio. Fluorescence lifetime measurements were based on the well-established time-correlated single-photon counting (TCSPC) technique.

3. Results and discussion

3.1. X-ray diffraction: phase purity and crystallite size

XRD patterns of Nd₂Zr₂O₇ (NZO) and Eu-doped Nd₂Zr₂O₇ (NEZO) are shown in Fig. 1.

Fig. 1 Powder X-ray diffraction pattern of as prepared Nd_2−xEu_xZr₂O₇. (a) x = 0, (b) x = 0.001, (c) x = 0.01, (d) x = 0.02, (e) x = 0.03 and (f) x = 0.05.

We have synthesized these compounds at 800 °C using the sol–gel combustion method, which guaranteed homogeneous mixing at the molecular level.

There are seven main identifiable diffraction peaks corresponding to Miller Indices (222), (400), (440), (622), (444), (800) and (662). They are in agreement with the reflections of a fluorite type structure (space group Fm3m).³¹ Both the compounds are indexed to a defect fluorite phase and europium doping has not distorted that basic fluorite network in Nd₂Zr₂O₇.

It is indeed reported that in the cases where Nd₂Zr₂O₇ samples are prepared in the temperature range of 700–1000 °C, they have a fluorite phase; whereas the one those calcined at 1100–1400 °C have a pyrochlore phase (space group Fd3m), which can be identified by the diffraction peak corresponding to (111) and (311), (331) and (511).³² There is no evidence of such diffraction peaks at 800 °C, thus indicating the absence of a pyrochlore phase in our prepared NZO and NEZO samples. The broadened XRD peak indicates the nanocrystalline nature of the as prepared un-doped and europium-doped neodymium zirconate. The crystallite size of the Nd₂Zr₂O₇ nanoparticle calculated from the Debye–Scherer formula is 7.1 ± 0.2 nm. Our detailed earlier report on blank Nd₂Zr₂O₇ also indicates that a sample prepared at 800 °C has a disordered fluorite phase as well as a nanocrystalline nature³³ that was confirmed using transmission electron microscopy and Fourier transformed infrared spectroscopy.

3.2. Photoluminescence properties

Fig. 2 shows the emission spectrum of the NZO sample under the excitation wavelength of 300 nm. The spectrum displays two broad features, viz. 450 nm and 535 nm, that correspond to blue and green emission, respectively. These emission bands cannot be attributed to Nd³⁺ because it is known to emit in the near infrared region. Also, the direct transition from conduction to valence band is not possible due to the very wide band gap of NZO. We assume that such a visible photoluminescence originates from the structural defects of the nanocrystals. Indeed, based on the density function theory (DFT) calculations in our earlier report, such emission in nanocrystalline Nd₂Zr₂O₇ is attributed to presence of singly ionized and doubly ionized oxygen vacancies.³³

Fig. 2 Emission spectrum of Nd₂Zr₂O₇ under excitation at 300 nm. Reproduced from ref. Gupta et al. Journal of Materials Chemistry C, 2016, DOI: 10.1039/C6TC01032F with permission from the Royal Society of Chemistry.

Defect-induced emission is very common for inorganic compounds ZrO₂, Al₂O₃, SiO₂, ZnO, In₂O₃, SnO₂ and CeO₂.³⁴ Many reports are available on the existence of defect-induced photoluminescence in zirconate-based perovskite-like lead zirconate,³⁵ barium zirconate,³⁶ calcium zirconate,³⁷ strontium zirconate,³⁸ etc. In all these cases, oxygen vacancy is the most common defect responsible for photoluminescence.

In this case, it is the presence of oxygen vacancies in the band gap of NZO that gives rise to visible bands. The blue band is because of a shallow level oxygen vacancy whereas the green band is because of similar defects at a deeper level. Zhang et al. have also observed oxygen vacancies related luminescence in RE₂Zr₂O₇ nanocrystals (RE = La, Nd, Eu and Y).³⁹

Fig. 3 shows the excitation spectrum of the 0.1 mol% Eu³⁺-doped Nd₂Zr₂O₇ sample under excitation at 611 nm corresponding to the ⁵D₀–⁷F₂ transition of Eu³⁺. Fig. S1 (ESI†) shows the dependence of excitation spectra on the concentration of europium ion. The broad band in the region 220–270 can be attributed to the charge transfer band (CTB). In general, CTB can arise because of contributions from various electronic transitions such as: (i) the host absorption band (HAB), which involves transfer of electrons from O → Zr(II) inter-valence electron transfer involving Eu(III) → Zr(IV) and (iii) a charge transfer state (CTS), which involves transfer of electron from O → Eu (major contribution). All these contributions together lead to such broadening in the CTB although the CTS contributes to the major fraction. Their contributions are difficult to resolve in complex oxide hosts such as A₂B₂O₇ due to spectral overlap. It is also observed that this CTB is red shifted with increasing europium ion doping concentration. Such phenomenon is assigned to an increase in covalency of the Eu–O bond and also to the local surrounding area around the europium ion, which further reduces the charge transfer energy.

Fig. 3 Excitation spectrum of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) under emission wavelength of 611 nm.

Based on the expression formulated by Jorgensen,⁴⁰ one can also predict charge transfer energy using eqn (1):

E^CT = 3.72(χ_X − χ_M) eV (1)

where χ_X is the optical electronegativity of the anion x from where the electron is transferred and χ_M is the optical electronegativity of the metal M where the electron is transferred.

For Nd₂Zr₂O₇ oxide based hosts, E^CT = 5.35 eV (∼232 nm), which is in the range predicted by experimental data. The exact value is tough to correlate due to the broad nature of the charge transfer in the excitation spectrum.

Upon careful examination of the excitation spectra, there are other peaks in the region 350–500 nm (magnified image in the inset of Fig. 3) that are attributed to the intra-configurational f–f transition of Eu³⁺. Among them, the peak at 395 nm is the most intense and is attributed to the ⁷F₀ → ⁵L₆ transition of the europium ion. Other peaks at 369, 381, 422, 442, 463 and 488 nm were assigned to the electronic transitions of ⁷F₀ → ⁵H₃, ⁵L₉, ⁵L₇, ⁵D₄, ⁵D₃,⁵D₂ and ⁵D₁, respectively. This indicates that this particular phosphor can be effectively excited by 395 nm (near UV) as well as 463 (blue LEDs). Fig. S1 (ESI†) shows the dependence of the excitation spectra on the concentration of europium ion.

The emission spectrum of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) under excitation at 256 nm is shown in Fig. 4a. The emission spectrum consisted of very weak emission from ⁵D₁ and ⁵D₂ levels, which is typical of a low phonon energy host. There are other relatively strong emission bands of ⁵D₀ → ⁷F₀ (575 nm), ⁵D₀ → ⁷F₁ (587 nm), ⁵D₀ → ⁷F₂ (611 nm), ⁵D₀ → ⁷F₃ (653 nm) and ⁵D₀ → ⁷F₄ (711 nm). Along with these, a substantial host emission due to oxygen-related defects could also be seen, which is an indication of incomplete energy transfer when 0.1 mol% of Eu³⁺ is doped into the Nd₂Zr₂O₇ host.

Fig. 4 (a) Emission spectrum of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) and (b) Stark splitting pattern of europium emission corresponding to the ⁵D₀ → ⁷F₀ transition under excitation at 256 nm.

Of all the transitions mentioned above, the ⁵D₀ → ⁷F₁ at 587 nm (ΔJ = ±1) and ⁵D₀ → ⁷F₂ at 615 nm (ΔJ = ±2) are categorized as site symmetry sensitive emission and referred to as magnetic dipole transition (MDT) and hypersensitive electric dipole transition (EDT), respectively. Although the more sensitive parameter used for structural probing is the ratio of (⁵D₀ → ⁷F₂) to (⁵D₀ → ⁷F₁), commonly known as the asymmetry ratio (I). Generally, in an inorganic lattice, if a particular site a center of inversion (C_i), EDT are strictly forbidden; whereas in a site without C_i, MDT is mostly absent and EDT is usually the strongest emission line because the transitions ΔJ = ±2 are hypersensitive to small deviations from inversion symmetry. The emission spectrum of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) under excitation at 395 nm is shown in Fig. S2 (ESI†). Upon excitation at λ = 395 nm, which corresponds to the ⁷F₀ → ⁵L₆ band, an emission spectrum similar to that for CTB excitation was observed, except that there is a large difference in intensities due to the fact that the f–f transition was forbidden in nature.

In the emission spectrum recorded at 256 nm excitation, the two most interesting features are: (i) intensity of EDT is more than MDT and (ii) the presence of highly intense ⁵D₀–⁷F₀ transition.

The fact that the intensity of ⁵D₀ → ⁷F₂ at 611 nm is more than ⁵D₀ → ⁷F₁ at 587 nm is an indication that the local symmetry around europium is low and deviated from an inversion center. In Nd₂Zr₂O₇, the coordination numbers of Nd and Zr are 8 and 6, respectively, with ionic radii of 111 and 72 pm, respectively. The ionic radii of the 8 and 6-coordinated trivalent europium ion are 107 and 95 pm, respectively. Eu³⁺ is much closer in size to Nd³⁺ and can easily occupy the larger Nd³⁺ site compared to very small sized Zr⁴⁺. Also, charge matching between Nd³⁺ and Eu³⁺ allow little distortion of the lattice. If the associated defect due to the charge/size difference is at a far-off distance, the local site symmetry around Eu³⁺ occupying Nd³⁺ will be a center of inversion where the smaller chunk occupying Zr⁴⁺ site will be without inversion symmetry. Thus, the observed emission spectrum where the ⁵D₀ → ⁷F₂ (EDT) at 611 nm transition of Eu³⁺ ions is much more intense than the ⁵D₀ → ⁷F₁ (MDT) transition at 587 nm can be attributed to the large fraction of Eu³⁺ ions occupying Zr⁴⁺ sites, though oxygen vacancies are introduced in the vicinity to ensure local charge compensation. The presence of MDT is an indication that some of the Eu³⁺ also occupies Nd³⁺ sites having inversion symmetry.

The fact that ⁷F₀ and ⁵D₀ levels are non-degenerate, since both the emitting and end states are non-degenerate, its number of components indicates the number of different metal–ion sites.⁴¹ The splitting in ⁵D₀ → ⁷F₀ indicates that an Eu³⁺ ion occupies two or more than non-equivalent sites.⁴² In the case of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%), the ⁵D₀ → ⁷F₀ levels do not exhibit a sharp and single peak. However, splitting with two components (Fig. 4b) is seen, which also justifies the emission spectroscopic results that Eu³⁺ is distributed into both Nd³⁺ and Zr⁴⁺ sites.

The other interesting observation is the very high intensity ⁵D₀ → ⁷F₀, even more than ⁵D₀ → ⁷F₁, which is rarely seen. Generally, the ⁵D₀ → ⁷F₀ transition of Eu³⁺. which is forbidden by both EDT as well as MDT mechanism, is an abnormal phenomenon is not in agreement with Judd–Ofelt theory.

Such abnormally intense ⁵D₀ → ⁷F₀ transitions have also been reported in other inorganic compounds^43–51 and different authors, including our group, have proposed different explanations for this. We proposed in our previous work on Sr₂SiO₄:Eu³⁺ that in a low symmetry environment, ⁵D₀ → ⁷F₀ steals some of the energy from ⁵D₀ → ⁷F₁ and ⁵D₀ → ⁷F₂. Fujishiro et al.^50,51 have proposed that it can be because of mixing of two different states with different J values or the breakdown of the Wybourne–Downer mechanism whereas Downer et al. has attributed such anomalies to relativistic effects.⁵² It is also reported that ⁵D₀ → ⁷F₀ normally appears when site symmetry around europium ion is very low and normally observed for following 10 site symmetries: C_s, C₁, C₂, C₃, C₄, C₆, C_2V, C_3V, C_4V and C_6V, according to the electric dipole selection rule.⁵³ It has also been reported for Y₃SbO₇:Eu³⁺. The presence of extra oxygen in the crystal lattice may sometimes cause unusually high intensity for ⁵D₀–⁷F₀ transitions.⁵⁴ In Nd₂Zr₂O₇, we believe that the presence of oxygen vacancies in the defect fluorite pyrochlore structure may induce the high transition probability for ⁵D₀–⁷F₀ even more than ⁵D₀–⁷F₁.

Fig. 5 schematically shows the excitation mechanism that occurs in Nd₂Zr₂O₇:Eu³⁺ upon 256 nm and 395 nm excitation, respectively. The excitation mechanism is indeed wavelength dependent. The first excitation mechanism at λ_ex = 256 nm is a CTS that involves transfer of electrons from O²⁻ → Eu³⁺ (major contribution), whereas the second one with λ_ex = 395 nm is due to the ⁷F₀ → ⁵L₆ transition of the europium ion (f–f transition). There is an additional process of de-excitation upon 256 nm radiation.

Fig. 5 Scheme of energy transfer pathways in Nd₂Zr₂O₇:Eu³⁺ upon (a) 256 nm and (b) 395 nm excitation.

One can determine the local site symmetry around the Eu³⁺ site based on the number of crystal-field splitting components in the ⁵D₀ → ⁷F_J transitions. In the case of Eu³⁺, ⁵D₀ → ⁷F₀, ⁵D₀ → ⁷F₁ (magnetic dipole transition) and ⁵D₀ → ⁷F₂ (hypersensitive electric dipole transition) are the ones that are the most affected by local site symmetry. ⁵D₀ → ⁷F₃ and ⁵D₀ → ⁷F₄ neither having a pure magnetic dipole nor having a pure electric dipole are still taken into consideration for the symmetry calculation. As discussed above, MDT is not perturbed much by the local environment whereas ⁵D₀–⁷F₂ (EDT) is. We have taken into consideration the spectral pattern of ⁵D₀ → ⁷F₀, ⁵D₀ → ⁷F₁ (MDT), ⁵D₀ → ⁷F₂ (EDT) ⁵D₀ → ⁷F₃ and ⁵D₀ → ⁷F₄ to arrive upon the point group symmetry of Eu³⁺ in Nd₂Zr₂O₇ at 0.1 mol% doping. Fig. 6a–d shows the slow scan recording of emission spectra of selective ⁵D₀ → ⁷F_1, ⁵D₀ → ⁷F₂, ⁵D₀ → ⁷F₃ and ⁵D₀ → ⁷F₄ regions.

Fig. 6 Stark splitting pattern of europium emission corresponding (a) ⁵D₀ → ⁷F₁ (b) ⁵D₀ → ⁷F₂ (c) ⁵D₀ → ⁷F₃ and (d) ⁵D₀ → ⁷F₄ under excitation wavelength of 256 nm.

The substitutions of Nd³⁺/Zr⁴⁺ with Eu³⁺ can results in substantial lattice distortions due to size mismatching at the Nd³⁺ site and both size and charge mismatching at the Zr⁴⁺ site. Stark components observed from local field splitting, as shown in Fig. 6a–d, two, two, three and five peaks for ⁵D₀ → ⁷F₁ (ΔJ = ±1), ⁵D₀ → ⁷F₂ (hypersensitive, ΔJ = ±2) ⁵D₀ → ⁷F₃ and ⁵D₀ → ⁷F₄ transition of Eu³⁺ were resolved for Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) respectively. According to the branching rules of various point groups,⁵⁵ the actual site symmetry of Eu³⁺ in Nd₂Zr₂O₇:Eu³⁺ reduces from the original D_3d for Nd³⁺/Zr³⁺ to C₄ is inferred.

3.3. Time resolved emission spectroscopy

The luminescence decay profile of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) is shown in Fig. 7. The measurements were carried out at excitation and emission wavelengths of 256 and 611 nm, respectively, and fitted using the following bi-exponential decay equation.where I(t) is the photoluminescence intensity, t is the time after excitation and τ_i is the decay life time for the i_th component having intensity I_i.

Fig. 7 Decay profile of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) under λ_ex = 256/λ_em = 611 nm.

Such multi-exponential decay behavior can arise due to various factors such as different luminescent centers, energy transfer processes, structural defects and the presence of impurities in the host.⁵⁶

Broadly, the lifetime fitting analysis showed the presence of three components: very short lived species (T₁ = 10.43 μs, 21%), medium range lifetime species (T₂ 1.14 ms, 54%) and long lived species (T₃ = 3.35 ms, 25%). T₁ is the life time of oxygen vacancy-related emission in the Nd₂Zr₂O₇ host, which is also evident from the presence of substantial broad emissions in the emission spectrum of the Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%). This is an indication of incomplete energy transfer from Nd₂Z_r2O₇ to Eu³⁺ in Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%).

If one considered the concept of phonon frequency, a longer lifetime species should be attributed to a more symmetric site, as the f–f transition becomes more forbidden, whereas the one with a shorter life time should be assigned to relatively asymmetric sites because of relaxation by the La-Porte selection rule. We have already discussed that Eu³⁺ at the 8-coordinated Nd³⁺ site has a more symmetric environment compared to the 6-coordinated Zr⁴⁺ site. The major Eu³⁺ species of T₂ (54%) has been attributed to Eu³⁺ in the Zr⁴⁺ site, whereas the minor fraction of T₃ (25%) is due to Eu³⁺ in the Nd³⁺ site. This is concordant with our emission spectral analysis where we have concluded that the majority of europium ions occupy Zr⁴⁺ sites whereas a minor fraction resides in Nd³⁺ sites as EDT dominates MDT. Such site selective spectroscopy of europium is done in various inorganic hosts such as strontium silicate, zinc pyrophosphate and strontium cerate where multiple sites are available for occupancy.^48,57,58

Based on emission and lifetime spectroscopy, it was inferred that the species for T₁ (10.43 μs, 21%) is due host emission, T₂ (1.14 ms, 54%) which is the major one, arises because of Eu³⁺ ions occupying the 6-coordinated Zr⁴⁺ site without inversion symmetry, whereas major species for T₃ (3.35 ms, 25%) can be attributed to Eu³⁺ ions occupying the 8-coordinated Nd³⁺ sites with inversion symmetry. In order to, identify the spectral features of T₁, T₂ and T₃, a detailed time resolved emission spectrometric (TRES) study was carried out on the system with λ_ex = 256 nm. Through this analysis, by giving suitable delay times and choosing a proper gate width (TRES data slicing range), the emission spectrum responsible for the particular decay time was obtained (Fig. 8) (a brief explanation regarding this is given in the ESI as I-1†). As can be seen from the spectra after delay of 50 μs; host emission due to oxygen vacancies could only be seen.

Fig. 8 Time-resolved emission spectra of Nd₂Zr₂O₇:Eu³⁺ (0.1 mol%) under the excitation at 256 and emission at 611 nm.

In the emission spectrum of T₂, the transition ⁵D₀ → ⁷F₀ (forbidden by both EDT and MDT) is very intense compared to ⁵D₀ → ⁷F₁ (MDT) and ⁵D₀ → ⁷F₂ (EDT) as well as a large splitting is found in ⁵D₀ → ⁷F₁ and ⁵D₀ → ⁷F₂ transitions. As was discussed earlier, the ⁵D₀ → ⁷F₀ transition of Eu³⁺ normally appears when site symmetry around the europium ion is very low and normally observed for the following 10 site symmetries: C_s, C₁, C₂, C₃, C₄, C₆, C_2V, C_3V, C_4V and C_6V. Also, the asymmetric ratio was found to be 1.58. This is an indication that this particular Eu³⁺ ion is present in a relatively asymmetric environment. From our earlier discussion, it was inferred that ZrO₆ is relatively less symmetric than NdO₈. Therefore, the T₂ species arises due to Eu³⁺ occupying the Zr⁴⁺ site.

In the emission spectrum of T₃, the transition ⁵D₀ → ⁷F₀ is of almost equal intensity to that of MDT and less intense to that of EDT. Also, in this case, the asymmetry ratio is less compared to that of T₂ (∼1.16). This is an indication of the fact that Eu³⁺ has reduced asymmetry in T₃. Hence, T₂ is attributed Eu³⁺ occupying the Nd³⁺ site.

3.4. Effect of Eu³⁺ ion concentration on luminescence properties of Nd₂Zr₂O₇:Eu

Emission spectra of Nd₂Zr₂O₇:Eu at different Eu³⁺ concentrations with an excitation wavelength of 256 nm is shown in Fig. 9. Interestingly, with increases in europium concentration, not only does an intensity change occur, but also the spectral features change immensely.

Fig. 9 (a) PL emission spectra (corrected w.r.t source) of the Nd₂Zr₂O₇:Eu³⁺ (mol% of Eu³⁺ = 0.1, 1.0, 2.0, 3.0 and 5.0) phosphors excited at 256 nm. (b) PL emission intensities of ⁵D₀ → ⁷F₂ as a function of Eu³⁺ ion concentration.

It can be very well seen from Fig. 9 that the emission intensity increases with increasing Eu³⁺ mol% yielding a maximum output at 3.0 mol%. Beyond 3.0%, emission intensity was found to decrease drastically, which is attributed due concentration quenching. This is supposedly the most predominant phenomenon taking place at higher dopant ion concentrations and is responsible for decreases in radiative emission by resonant energy transfer between Eu³⁺ ions. To have better insight into the exact mechanism responsible for quenching, the critical distance (r_c) needs to be evaluated using eqn (3):⁵⁹

where r_c is the critical distance between the dopant ion and quenching site, V is the volume of the unit cell, X_c is the critical concentration of Eu³⁺ ions and N is the number of cations per unit cell.

For the Nd₂Zr₂O₇ host lattice, V = 1216.82 ų, N = 8 (four formula units per one unit cell) and X_c = 0.03. Based on these values, the critical distance is calculated to be 13.64 Å. In this case, the Eu³⁺–Eu³⁺ distance is larger than 10 Å. Therefore, concentration quenching due to exchange interactions is obsolete. Therefore, the electric multipolar interaction is believed to be the main mode of non-radiative energy transfer among the Eu³⁺ ions and is responsible for concentration quenching in the Nd₂Zr₂O₇ phosphor.

If one carefully observes the concentration dependent emission profile, it can be very well seen that at all concentration EDT predominated MDT, which indicated that the majority of europium ions have asymmetric local surroundings. Although the main point of interest is how the intensity of the ⁵D₀ → ⁷F₀ transition changes as a function of europium ion concentration. From Fig. 7, it can be seen that ⁵D₀ → ⁷F₀ was more intense than ⁵D₀ → ⁷F₁ at 0.1 mol% Eu³⁺ and this trends seems to reverse at higher concentrations. As the europium concentration increases, the intensity of ⁵D₀ → ⁷F₀ starts decreasing, almost equalizing at 3.0 mol%, and interestingly at 5.0 mol%, ⁵D₀ → ⁷F₁ is more than that of ⁵D₀ → ⁷F₀. This phenomenon can be attributed to a reduction in the concentration of oxygen vacancies at higher Eu³⁺ concentrations.

Another interesting observation is that as the concentration of europium ion increases, host to dopant energy transfer efficiency increases and finally complete transfer is observed in the case of 5.0 mol% doping. This is fully supported by lifetime spectroscopy data mentioned in Table 1. It can be very well seen from this particular table that the T₁ species (host emission) lifetime remains at all europium concentrations, but its population reduces as the Eu³⁺ concentration increases and disappears completely at 5.0 mol%, which displays bi-exponential behavior Fig. S3 (ESI†) indicating complete energy transfer at the 5% doping level. The fact that the population of T₁ gradually recedes as the europium concentration increases is indicative of the fact that host-dopant energy transfer is becoming more efficient at higher concentrations. It can also be seen from the lifetime value of T₂ and T₃ which is due to europium ion emission that as the dopant concentration increases, the lifetime value also increases upto 3.0 mol% but decreases at 5.0 mol% due to concentration quenching. This is in complete concordance with trends observed in excitation and emission spectroscopy.

Table 1 Lifetime value and their relative population of Nd₂Zr₂O₇:Eu at different europium ion concentrations

Concentration of Eu^3+ (mol%)	T1 (μs)	T2 (ms)	T3 (ms)
0.1	10.43 (21%)	1.14 (54%)	3.35 (25%)
1.0	10.63 (16%)	1.32 (58%)	3.56 (26%)
2.0	10.85 (12%)	1.55 (60%)	3.75 (28%)
3.0	11.02 (8%)	1.69 (64%)	3.85 (28%)
5.0	—	1.02 (61%)	2.96 (39%)

---

3.5. Photophysical characteristics and colorimetric performance

Based on the emission spectra (corrected w.r.t source, monochromator and detector) of europium-doped inorganic phosphors, one can evaluate various optical properties such as radiative and non-radiative lifetimes, quantum yield, branching ratio, etc. For lanthanide ions other than Eu³⁺; there is no pure MDT and therefore, one needs absorption spectrum data unlike the europium ion for such analyses. The Detailed procedure for all these calculations is mentioned in our earlier report.⁵⁷ Other than this, one of the most important optical parameters are the Judd–Ofelt (JO) intensity parameters, Ω_J (J = 2, 4) which are calculated based on classical Judd–Ofelt theory.^60,61 This classical parameter gives information about the local symmetry, covalency and polarizability around the lanthanide ion. The short range parameter, Ω₂, gives information about the local properties such as degree of covalency and polarizability of the chemical environment experienced by the Ln³⁺ ion in an inorganic host. On the other hand, the long range parameter, Ω₄, gives information about the bulk properties such as viscosity and rigidity of the inorganic crystal.⁵⁷

For Nd₂Zr₂O₇, we adopted the refractive index value of 2.11. The JO parameter and other photophysical values are mentioned in Table 2.

Table 2 Photophysical properties Eu³⁺-doped Nd₂Zr₂O₆

Mol%	ARAD (s^−1)	ANRAD (s^−1)	η (%)	Ω2 (×10^−20 cm^2)	Ω4 (×10^−20 cm^2)
0.01	425	89	79.8	2.09	0.769
1.0	462	81	82.2	1.47	0.521
2.0	469	77	85.9	1.79	0.634
3.0	496	63	89.0	2.09	0.738
5.0	240	366	39.5	1.45	0.748

---

At all concentrations, the Ω₂ value is greater than the Ω_4, indicating the existence of high covalency between Eu³⁺ and ligand (Eu–O), and a relatively asymmetric environment around the Eu³⁺ ion in Nd₂Zr₂O₇, which is highly probable because it is distributed between 8-coordinated Nd³⁺ and 6-coordinated Zr⁴⁺ (size and charge imbalance). This is also reflected in the intense EDT compared to MDT and the observation of an intense ⁵D₀ → ⁷F₀ transition in our emission spectra that is forbidden by both EDT as well as MDT modes.

From Table 2, with an increase in Eu³⁺ ion concentration up to 2.0 mol%, the radiative transition rate (A_R) increases and the non-radiative rate (A_NR) decreases due to more incorporation of luminescence active europium ions. This is in complete concordance with our emission spectral data where concentration quenching was observed beyond 3.0 mol% due to non-radiative energy transfer between closely lying europium ions. This is also accordingly reflected in another important parameter, the quantum efficiency, which is highest for the 3.0 mol% sample with a value of 89%.

To qualify the material as a good phosphor, colorimetric performance is one of the most important properties that need to be evaluated. CIE chromaticity coordinates were evaluated for the Nd₂Zr₂O₇:Eu³⁺ (3.0 mol%) sample using standard protocols from the corrected emission spectrum. This is represented as the point ‘#’ in the CIE diagram shown in Fig. 10. It is clear from the CIE index (0.614, 312) that it is closer to the standard value (0.67 and 0.32) and that Nd₂Zr₂O₇:Eu³⁺ gives an intense red emission with a very high quantum efficiency of 89. This clearly demonstrates its applicability as a promising red phosphor under near UV for white light emitting diodes.

Fig. 10 CIE diagram showing the coordinates and representing the color emitted by Nd₂Zr₂O₇:Eu³⁺ (3.0 mol% doping).

4. Conclusion

Varied concentrations of Eu³⁺-doped Nd₂Zr₂O₇ were synthesized using the gel-combustion method. The material was systematically characterized using X-ray diffraction (XRD) and time resolved photoluminescence spectroscopy (TRPLS). On irradiating the un-doped Nd₂Zr₂O₇, it emits visible light, which was attributed to the presence of oxygen vacancies within its band gap. On doping with europium, three interesting observations take place: (i) incomplete host dopant energy transfer (ii) unusually high intensity of ⁵D₀ → ⁷F₀ and (iii) high crystal field splitting. Based on emission characteristics, dual splitting in the ⁵D₀ → ⁷F₀ transition and lifetime spectroscopy; it was found that europium occupies both Nd³⁺ sites with a lifetime value of 3.34 ms as well as Zr⁴⁺ sites with a lifetime value of 1.14 ms. The population analysis using emission kinetics shows that more europium is localized at the Zr⁴⁺ site compared to the Nd³⁺ site. The un-doped sample showed visible emission due to the presence of oxygen vacancies. An unusually high intensity of the ⁵D₀ → ⁷F₀ transition can be attributed to the presence of oxygen vacancy-related defects. The low phonon frequency of the Nd₂Zr₂O₇ host was confirmed from transition from ⁵D₁ and ⁵D₂ levels along with standard ⁵D₀ → ⁷F_J Eu³⁺ emission lines. As the doping concentration of europium ion increases, the intensity of the ⁵D₀ → ⁷F₀ was found to gradually decrease due to compensation of oxygen vacancies by the high level of europium doping. Interestingly, as the europium percentage increases, the efficiency of host-dopant energy transfer also increases, and indeed, complete host-dopant energy transfer takes place at 5.0% Eu³⁺ doping. This is also reflected by the bi-exponential decay behavior at 5.0 mol% Eu³⁺. Based on time resolved emission spectroscopy (TRES), the emission spectrum of europium at the Nd³⁺ site, as well as at the Zr⁴⁺ site, was elucidated. Calculation of Judd–Ofelt parameter illustrated that at all europium concentrations, the local surrounding environment lacks inversion symmetry (asymmetric environment) and the exact point group symmetry was found to be C₄. The red purity and high quantum efficiency of Nd₂Zr₂O₇:Eu³⁺ projects it as a new material for commercial application in white light emitting diodes.

References

1. S. K. Gupta, P. S. Ghosh, A. K. Yadav, N. Pathak, S. N. Jha, A. Arya, D. Bhattacharyya and R. M. Kadam, Inorg. Chem., 2016, 55, 1728.
2. S. K. Gupta, K. Sudarshan, P. S. Ghosh, S. Mukherjee and R. M. Kadam, J. Phys. Chem. C, 2016, 120, 7265.
3. S. K. Gupta, P. S. Ghosh, K. Sudarshan, R. Gupta, P. K. Pujari and R. M. Kadam, Dalton Trans., 2015, 44, 19097.
4. J. Ren, G. Jia, Y. Guo, A. Wang and S. Xu, J. Phys. Chem. C, 2016, 120, 1342.
5. M. Wang, H. Ye, L. You and X. Chen, ACS Appl. Mater. Interfaces, 2016, 8, 574.
6. A. I. Becerro, D. G. Mancebo, E. Cantelar, F. Cussó, G. Stepien, J. M. D. L. Fuente and M. Ocaña, Langmuir, 2016, 32, 411.
7. S. M. Tehrani, W. Lin, S. Rosenfeldt, G. Guerin, Y. Lu, Y. Liang, M. Drechsler, S. Forster and M. A. Winnik, Chem. Mater., 2016, 28, 501.
8. S. K. Gupta, M. Sahu, P. S. Ghosh, D. Tyagi, M. K. Saxena and R. Kadam, Dalton Trans., 2015, 44, 18957.
9. R. Phatak, S. K. Gupta, K. Krishnan, S. K. Sali, S. V. Godbole and A. Das, Dalton Trans., 2014, 43, 3306.
10. S. K. Gupta, P. S. Ghosh, M. Sahu, K. Bhattacharyya, R. Tewari and V. Natarajan, RSC Adv., 2015, 5, 58832.
11. B. G. Vats, S. K. Gupta, M. Keskar, R. Phatak, S. Mukherjee and S. Kannan, New J. Chem., 2016, 40, 1799.
12. S. K. Gupta, B. Rajeshwari, S. N. Achary, S. J. Patwe, A. K. Tyagi, V. Natarajan and R. M. Kadam, Eur. J. Inorg. Chem., 2015, 2015, 4429.
13. S. K. Gupta, M. Sahu, K. Krishnan, M. Saxena, V. Natarajan and S. V. Godbole, J. Mater. Chem. C, 2013, 1, 7054.
14. (a) N. Pathak, P. S. Ghosh, S. K. Gupta, S. Mukherjee, R. M. Kadam and A. Arya, J. Phys. Chem. C, 2016, 120, 4016; (b) N. Pathak, S. K. Gupta, K. Sanyal, M. Kumar, R. M. Kadam and V. Natarajan, Dalton Trans., 2014, 43, 9313.
15. R. Shukla, S. K. Gupta, V. Grover, V. Natarajan and A. K. Tyagi, Dalton Trans., 2015, 44, 10628.
16. S. K. Gupta, K. Sudarshan, P. S. Ghosh, K. Sanyal, A. P. Srivastava, A. Arya, P. K. Pujari and R. M. Kadam, RSC Adv., 2016, 6, 3792.
17. S. K. Gupta, S. Nigam, A. K. Yadav, M. Mohapatra, S. N. Jha, C. Majumder and D. Bhattacharyya, New J. Chem., 2015, 39, 6531.
18. M. A. Subramanian, G. Aravamudan and G. V. S. Rao, Prog. Solid State Chem., 1983, 15, 55.
19. E. J. Harvey, K. R. Whittle, G. R. Lumpkin, R. I. Smith and S. A. T. Redfern, J. Solid State Chem., 2005, 178, 800.
20. Y. Tong, J. Zhu, L. Lu, X. Wang and X. Yang, J. Alloys Compd., 2008, 465, 280.
21. (a) J. Xu, V. K. Anand, A. K. Bera, M. Frontzek, D. L. Abernathy, N. Casati, K. Siemensmeyer and B. Lake, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 92, 224430; (b) E. Lhotel, S. Petit, S. Guitteny, O. Florea, M. C. Hatnean, C. Colin, E. Ressouche, M. R. Lees and G. Balakrishna, Phys. Rev. Lett., 2015, 115, 197202.
22. S. Solomon, A. George, J. K. Thomas and A. John, J. Electron. Mater., 2015, 44, 28.
23. C. Fisher, S. Finkeldei, F. Brandt, D. Bosbach and A. Luttage, ACS Appl. Mater. Interfaces, 2015, 7, 17857.
24. B. V. Kumar, R. Velchuri, G. Prasad, B. Sreedhar, K. Ravikumar and M. Vithal, Ceram. Int., 2010, 36, 1347.
25. T. Feng, D. R. Clarke, D. Jiang, J. Xia and J. Shi, Appl. Phys. Lett., 2011, 98, 151105.
26. S. K. Gupta, V. Grover, R. Shukla, K. Srinivasu, V. Natarajan and A. K. Tyagi, Chem. Eng. J., 2016, 283, 126.
27. S. K. Gupta, M. Mohapatra, V. Natarajan and S. V. Godbole, RSC Adv., 2013, 3, 20046.
28. S. K. Gupta, M. Mohapatra, V. Natarajan and S. V. Godbole, J. Mater. Sci., 2012, 47, 3504.
29. C. M. Wang, K. S. Kao, S. Y. Lin, Y. C. Chen and S. C. Wang, J. Phys. Chem. Solids, 2008, 69, 608.
30. http://www.ethesis.nitrkl.ac.in/1526/1/sp_thesis.pdf.
31. L. Kong, I. Karatchevtseva, D. J. Gregg, M. G. Blackford, R. Holmes and G. Triani, J. Eur. Ceram. Soc., 2013, 33, 3273.
32. S. K. Gupta, K. Sudarshan, P. S. Ghosh, A. P. Srivastava, S. Bevara, P. K. Pujari and R. M. Kadam, J. Mater. Chem. C, 2016 10.1039/c6tc01032f.
33. Y. H. Lee, H. S. Sheu and H.-C. I. Kao, Mater. Chem. Phys., 2010, 124, 145.
34. C. Zhang and J. Lin, Chem. Rev., 2012, 41, 7938.
35. J. M. A. Nunes, J. W. M. Espinosa, M. F. C. Gurgel, P. S. Pizani, S. H. Leal, M. R. M. C. Santos and E. Longo, Ceram. Int., 2012, 38, 4593.
36. S. K. Gupta, N. Pathak and R. M. Kadam, J. Lumin., 2016, 169, 106.
37. S. K. Gupta, P. S. Ghosh, N. Pathak and R. Tewari, RSC Adv., 2015, 5, 56526.
38. V. M. Longo, L. S. Cavalcante, R. Erlo, V. R. Mastelaro, A. T. De Figueiredo, J. R. Sambrano, S. de Lazaro, A. Z. Freitas, L. Gomes, N. D. Vieira Jr., J. A. Varela and E. Longo, Acta Mater., 2008, 56, 2191.
39. A. Y. Zhang, M. K. Lu, Z. S. Yang, G. J. Zhou and Y. Y. Zhou, Solid State Sci., 2008, 10, 74.
40. C. K. Jorgensen, Modern Aspects of Ligand Field Theory, North-Holland Publishing Company, Amsterdam, 1971.
41. F. Du, R. Zhu, Y. L. Huang, Y. Tao and H. J. Seo, Dalton Trans., 2011, 40, 11433.
42. K. Binnemans and C. Gorller Walrand, J. Rare Earths, 1996, 14, 173.
43. T. S. Sreena, P. P. Rao, T. L. Francis, A. K. V. Raj and P. S. Babu, Dalton Trans., 2015, 44, 8818.
44. L. Smentek and A. Kedziorski, J. Alloys Compd., 2009, 488, 586.
45. A. S. Souza, Y. A. R. Oliveira and M. A. Couto dos Santos, Opt. Mater., 2013, 35, 1633.
46. A. M. Pires, M. R. Davolos and O. L. Malta, J. Lumin., 1997, 72, 244.
47. B. Piriou, M. Richard-Plouet, J. Parmentier, F. Ferey and S. Vilminot, J. Alloys Compd., 1997, 262, 450.
48. N. Lakshminarasimhan and U. V. Varadaraju, J. Solid State Chem., 2004, 177, 3536.
49. S. K. Gupta, M. Mohapatra, S. Kaity, V. Natarajan and S. V. Godbole, J. Lumin., 2012, 132, 1329.
50. F. Fujishiro, R. Sekimoto and T. Hashimoto, J. Lumin., 2013, 133, 217.
51. F. Fujishiro, M. Murakami, T. Hashimoto and M. Takahashi, J. Ceram. Soc. Jpn., 2010, 118, 1217.
52. M. C. Downer, G. W. Burdick and D. K. Sardar, J. Chem. Phys., 1988, 89, 1787.
53. S. K. Gupta, M. K. Bhide, S. V. Godbole and V. Natarajan, J. Am. Ceram. Soc., 2014, 97, 3694.
54. J. Wang, Y. Cheng, Y. Huang, P. Cai, S. I. Kim and H. J. Seo, J. Mater. Chem. C, 2014, 2, 5559.
55. K. Binnemans, Coord. Chem. Rev., 2015, 295, 1.
56. G. Jia, Y. Song, M. Yang, Y. Huang, L. Zhang and H. You, Opt. Mater., 2009, 31, 1032.
57. S. K. Gupta, M. Mohapatra, V. Natarajan and S. V. Godbole, RSC Adv., 2013, 3, 20046.
58. S. K. Gupta, M. Sahu, K. Krishnan, M. Saxena, V. Natarajan and S. V. Godbole, J. Mater. Chem. C, 2013, 1, 7054.
59. G. Blasse, Phys. Lett. A, 1968, 28, 444.
60. B. R. Judd, Phys. Rev., 1962, 127, 750.
61. G. S. Ofelt, J. Chem. Phys., 1962, 37, 511.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra11698a

---