Investigation of defect related photoluminescence property of multicolour emitting Gd₂O₃:Dy³⁺ phosphor

Abstract

Multicolour emitting Gd₂O₃:Dy³⁺ phosphor is prepared by citrate based sol–gel method as a function of annealing temperature of the sample and its emission property dependence with lattice defects is elaborated. The annealed phosphors are characterized by X-ray powder diffraction (XRD), Raman spectroscopy, diffuse reflectance spectroscopy (DRS), photoluminescence (PL), fluorescence lifetime and positron annihilation lifetime spectroscopy (PALS). The phosphors annealed at different temperatures greatly influence the defect and emission intensities, as revealed from PL and PALS measurements, respectively. The efficient energy transfer (ET) from the Gd³⁺ ion to Dy³⁺ ion is schematically illustrated with the aid of an energy level diagram. The PL spectra clearly conclude that Dy³⁺ with an ionic radius close to the Gd³⁺ ion prefers to occupy the C₂ site in the Gd₂O₃ matrix. The positron lifetime spectroscopy qualitatively explains the concentration of defects (vacancy and voids) which are minimized with the increase in the annealing temperature. The correlation between PL emission and lattice defects is also reported in detail.

---

Introduction

Luminescent materials have attracted a great deal of interest due to their wide applications in various fields such as LEDs, biological assays, display and photonic devices.^1–5 Among them, much focus is given to the LED based on white light emission owing to its long photoluminescence lifetime, high brightness, low power consumption and environmental friendliness (i.e. mercury free) and it is considered as the next generation of solid state lightening devices.^6–8 The efficiency of LEDs is also found to be higher than the conventionally used incandescent and fluorescent light sources.⁹

Among the rare earth (RE) ions, Dy³⁺ provides strong blue, yellow and feeble red emission in the visible region of the electromagnetic spectrum due to the 4f–4f transition. Various colour emissions are possible due to the transition from the excited state of ⁴F_9/2 to ⁶H_15/2 (blue), ⁶H_13/2 (yellow), ⁶H_11/2 (red) state.^10–13 Based on the occupancy of Dy³⁺ ion in the inversion or non-inversion site of Gd³⁺, the intensities of emission colours differ.¹⁴ By tuning and coupling the emission intensity of each colour, it is possible to obtain white colour emission. When compared to direct UV excitation of Dy³⁺, the energy transfer through host ion or sensitizer ion is more efficient.¹³ Recently an effective energy transfer has been reported from Gd³⁺ to Dy³⁺ in the Ca₂Gd₈Si₆O₂₆:Dy³⁺ and Gd(BO₂)₃:Dy³⁺ phosphor.^15,16

In addition, the luminescence properties are strongly influenced by the presence of lattice defects¹⁷ like mono vacancy, vacancy cluster, voids or pores which are formed naturally during synthesis. The positron annihilation spectroscopy is a powerful and sensitive tool to identify the defects in the solids, distinguish the type of defects and provide relevant information about the size and density around the defects.¹⁸ The positively charged positrons are generated by an unstable isotope, typically ²²Na, which is marked by the simultaneous release of a 1.28 MeV gamma rays. On entering into the condensed matter, the positron searches for a negatively charged electron and annihilates with the release of two 511 keV γ quanta. The positively charged positron is easily attracted towards the vacancy and open space in order to neutralize the charge. The lifetimes measured for the positrons trapped in a certain defect type are sensitive to the defect size. The value determines the type of defects¹⁹ and the corresponding intensity reveals the defect concentration. An inverse relation exists between the positron lifetime and the electron density of the trap site. The energy distribution of the annihilation γ rays is broadened by the momentum component of the annihilating electron–positron pair p_L which is parallel to the emitting direction of the γ rays. The energy of the γ rays is given by E_γ = 511 ± ΔE_γ keV. Here, the Doppler shift ΔE_γ is given by ΔE_γ = p_Lc/2, where c is the speed of light. A freely diffusing positron may be localized in a vacancy-type defect because of Coulombic repulsion from the positively charged ion cores. Because the momentum distribution of the electrons in such defects differs from that of electrons in the bulk material, these defects can be detected by measuring the Doppler broadening spectra of the annihilation radiation. From the above, the S parameter represents the fraction of low momentum electrons annihilated by the positrons and, hence, it varies in accordance with the changes in size and concentration of the defects in the solids.²⁰

Lu et al. reported the positron lifetime of the Eu³⁺/Dy³⁺ codoped SrAl₂Si₂O₈/SrAlSi_1/2O_7/2 composite phosphor.²¹ The lattice defect created by replacing Sr by Eu and Dy in the composite phosphor affects the photoluminescence emission intensity. Defect properties of La and Eu codoped CaAl₂O₄ phosphor is investigated by Lin et al.²² Here, the defects due to the charge imbalance between La³⁺ and Ca²⁺ is focused and the defect concentration increases with increasing the content of La. The formation of vacancy clusters within the nanoparticles is reported by Thorat et al. in Eu doped CeO₂.²³ On increasing the positive charge defect by doping Er³⁺ and Nd³⁺, the photoluminescence emission intensity of BaAlMg₁₀O₁₇:Eu²⁺ is enhanced.²⁴ To the best of our knowledge, the positron annihilation studies on Gd₂O₃:Dy³⁺ phosphors are not explored elsewhere.

In the present work, Gd₂O₃:Dy³⁺ phosphors are synthesized by citrate-based sol–gel method and it is annealed at different temperatures. The photoluminescence properties of Dy³⁺ ion are investigated through energy transfer mechanism which is initiated from Gd³⁺ to Dy³⁺ and the process is well illustrated by the energy level diagram. The combined emission colour is also predicted and integrated from the CIE coordinate. Since the photoluminescence emission property strongly depends on the concentration of impurities and defects, we employed a high sensitive positron annihilation spectroscopy to identify and to investigate the defects in Gd₂O₃:Dy³⁺ phosphors. Finally, we also explore the correlation between the lattice defect and photoluminescence properties.

Experimental and characterization techniques

0.1 M of citric acid is dissolved in de-ionized water and used as a solvent. The required quantity of Gd(NO₃)₃ and Dy(NO₃)₃ are dissolved in citric acid with magnetic stirring to get homogeneous solution. The molar ratio of the metal ions to the citric acid is maintained as 1 : 1. The pH of the solution is raised to 2 by the adding aqueous NH₃ solution in drops. The solution is heated to 70 °C with contant stirring for the solvent evaporation. Initially, a transparent gel is formed and gradually transformed to xerogel. The residual organic precursor present in the sample is removed by heating at 250 °C for 2 h which yield a black-coloured fine particles. The Gd_1.99O₃:Dy³⁺_0.01 phosphors are obtained by annealing at 600 °C, 700 °C and 800 °C in air atmosphere.

The crystal structure and lattice parameters are estimated from Rigaku Ultima III powder X-ray diffractometer (XRD). The pattern is recorded at θ–2θ geometry with CuKα₁ (1.5406 Å) generated at 40 kV and 30 mA over the range of 10° to 80° with a step size of 0.02°. Transmission electron microscopy (TEM) images are recorded using Tecnai G² FEI with the accelerating voltage of 200 kV. For TEM measurements, the powder samples are dispersed well in ethanol and a drop casted on carbon-coated copper grid. Fourier transform infrared (FTIR) spectra are recorded between 4000 and 400 cm⁻¹ using Jasco FTIR-4200. Raman spectra are recorded using Jobin Yvon Horiba (LABRAM HR - 800) Micro Raman spectrometer attached with Nd:YAG laser as the excitation source having an output power of 15 mW with a laser beam spot size 100 mm in the range of 700 cm⁻¹ to 300 cm⁻¹. The absorption property of the material is measured with diffuse reflectance spectroscopy (DRS) using UV 2600 Shimadzu instruments. The photoluminescence (PL) spectra are recorded with spectroflurometer (RF 5301 PC Shimadzu) equipped with 150 W Xe source. Quantum yield is determined using JASCO ISF – 834 60 mm integrating sphere at the incident angle of 5°. Fluorescence lifetime measurements are obtained under pulsed excitation at 266 nm by using Nd:YAG laser with a pulse width of 8 ns and pulse energy of 150 mJ. The positron annihilation lifetime spectroscopy (PALS) is measured using a conventional lifetime system with digital oscilloscope (LeCroy Wavepro). The positron annihilation signals are detected by two BaF₂ scintillators attached to H3378 (Hamamatsu Photonics) photomultiplier tubes (PMT). The output of PMT is divided into two: one is connected to a constant fraction discrimintor (CFD) and the other to the digital oscilloscope. The samples are sandwiched with ²²Na source (370 kBq) and lifetime measurements are carried out in room atmospheric pressure. Each spectrum has counts about 6 x10⁶ and the number of lifetime components is determined by deconvoluting the lifetime spectra. The observed spectra are analysed and interpreted with a time resolution of about 170 ps by using the RESOLUTION computer program.²⁵ Coincidence Doppler broadening spectra are measured by using conventional spectrometer with Ge detector, with energy resolution of ∼1.2 keV at the photo peak of 511 keV. For each of the measurements, a total of 5 × 10⁶ counts in the annihilation peak are collected. The low-momentum part of the spectra is characterized by the S parameter, defined as the number of annihilation events over the energy range of 511 keV ± ΔE_γ (where ΔE_γ = 0.76 keV) around the centre of the peak, and the high-momentum part is characterized by using the W parameter, defined as the annihilation events in the range of 3.4 keV ≤ ∣ΔE_γ∣ ≤ 6.8 keV.

Results and discussion

The phase purity of the phosphors is analysed by XRD pattern. Fig. 1 shows the XRD pattern of Gd₂O₃:Dy³⁺ annealed at different temperatures. All the diffraction peaks of the phosphors are indexed to cubic phase Gd₂O₃ (JCPDS card no. 86 2477) with the space group of Ia3̄ (206) and lattice parameter of a = b = c = 10.8 Å. The most prominent peaks in the XRD pattern (Fig. 1) are indexed and the corresponding (hkl) planes are indicated. The absence of secondary X-ray diffraction peak related to the Dy₂O₃ phase confirms that the Dy³⁺ ions are well substituted in the Gd³⁺ site without influencing the crystal structure of Gd₂O₃. Moreover, dissolving 1% of Dy³⁺ in Gd₂O₃ lattice is highly possible due to their similar ionic radii (Gd³⁺ = 0.938 Å and Dy³⁺ = 0.912 Å). In Gd₂O₃:Dy³⁺ annealed at 600 °C, peaks in the 2θ range of 50° and 80° are not well observed but above 600 °C annealed samples, all the peaks are clearly distinguishable revealing the high crystalline nature of the phosphor.

Fig. 1 XRD pattern of Gd₂O₃:Dy³⁺ phosphors annealed at different temperatures.

The peak broadening in the XRD pattern due to the crystallite size is calculated after eliminating the instrumental broadening. The contribution of the instrumental effect is eliminated using Cauchy's relation²⁶ presented in eqn (1).

β_hkl = (β_hkl)_m − (β_hkl)_i (1)

where (β_hkl)_m and (β_hkl)_i are the full width half maximum (FWHM) of the measured XRD peak and the standard material such as silicon respectively. The interplanar distance (d_hkl) and the average crystallite size (D) are estimated using¹⁹ eqn (2) and (3).

2d_hkl sin θ = nλ (2)

where λ is the wavelength of X-ray used, θ is Bragg's angle and n is an integer. The average crystallite size and the lattice parameters of the phosphors with respect to (222) plane are calculated and tabulated in Table 1. When Gd₂O₃:Dy³⁺ is annealed at 600 °C, the calculated D value is 15 nm and so the surface to volume ratio is high and more number of Dy³⁺ ions may lie on the surface. Further, on annealing the phosphors to higher temperatures, the D value also increases and the Dy³⁺ ions on the surface starts diffusing and migrating towards the symmetric sites of Gd₂O₃. The substitution of Dy³⁺ is clearly evident from the variation of lattice parameter and shift in 2θ peak position in comparison to the standard Gd₂O₃ (JCPDS-86-2477). The increase in the average crystallite size is due to the self-nucleation of nanocrystallite material during the annealing process.²⁷

Table 1 Lattice parameters, crystallite size and CIE coordinated of Gd₂O₃:Dy³⁺ phosphors

Annealing temperature	2θ (°)	β (rad) ×10^−3	dhkl (Å)	D (nm)	a = b = c (Å)	Unit cell volume (Å^3)
JCPDS-86 2477	28.585	—	3.120	—	10.8	1259.71
600 °C	28.636	9.961	3.116	15	10.795	1257.96
700 °C	28.561	4.71	3.124	33	10.823	1267.77
800 °C	28.464	3.332	3.135	47	10.859	1280.47

---

Fig. 2(a) shows the TEM micrograph of Gd₂O₃:Dy³⁺ phosphor annealed at 600 °C. We observe that the phosphors exhibit more or less spherical shape morphology with non-uniform size distribution. The particle size distribution presented in Fig. 2(b) implies that the size of nanoparticles is in the range of 9–16 nm which is in consistent with the size (15 nm) estimated from XRD pattern. Fig. 2(c) shows the selected area diffraction (SAD) pattern of Gd₂O₃:Dy³⁺ phosphor, the presence of ring-like pattern confirms the polycrystalline nature and cubic structure of Gd₂O₃. The rings of SAD pattern are indexed to the corresponding plane.

Fig. 2 (a) TEM micrograph (b) size distribution graph (c) selected area diffraction pattern of Gd₂O₃:Dy³⁺ annealed at 600 °C.

The presence of various chemical functional groups in the phosphor is studied from the FTIR spectra as shown in Fig. 3. The broad peaks centred at 3400 cm⁻¹ corresponds to the O–H stretching of H₂O molecule absorbed from the atmospheric air.²⁸ However, this peak starts diminishing and completely disappeared on increasing the annealing temperature. The weak intense peak around 1600 cm⁻¹ indicate the vibration of C=O group which originate from the absorption of atmospheric CO₂.²⁹ The absence of vibration peaks corresponding to the citric acid functional groups (in the range of 1730 cm⁻¹–1700 cm⁻¹ and 1710 cm⁻¹–1680 cm⁻¹) implies that the annealing effectively removes the solvent medium and render a phase pure Gd₂O₃.²⁸ The sharp peak appearing at 545 cm⁻¹ is attributed to the vibration of Gd–O bond³⁰ and it becomes more intense and predominant with the increase in annealing temperature.

Fig. 3 FTIR spectra of Gd₂O₃:Dy³⁺ annealed at different temperatures.

The phonon energy of the Gd₂O₃:Dy³⁺ phosphors annealed at 600 °C, 700 °C and 800 °C phosphor is characterized by Raman spectroscopy and presented in Fig. 4. The spectra consists of Raman mode of vibrations at 359 cm⁻¹, 314 cm⁻¹, 445 cm⁻¹, 479 cm⁻¹ and 568 cm⁻¹.^31–33 When the phosphors are annealed from 600 °C to 800 °C, crystallinity nature increases and so the sharpness of the Raman peak. In order to avoid, quenching of excited energy due to non-radiative transition, it is necessary to select host matrix with low phonon energy.³⁴ The phonon energy corresponding to the maximum intensity peak (359 cm⁻¹) is not sufficient enough to undergo transition in electronic energy level. It clearly indicates that Gd₂O₃ is a host matrix with low phonon energy and hence a promising host for luminescent applications.

Fig. 4 Raman spectra of Gd₂O₃:Dy³⁺ phosphor annealed at 600 °C, 700 °C and 800 °C.

Fig. 5(a) shows the reflectance spectra of Gd₂O₃:Dy³⁺ phosphors annealed at different temperatures and the wavelength is recorded between 200 nm and 800 nm. It shows strong host ion absorption between 200 nm and 250 nm with the sharp adsorption edge at 231 nm. The weak adsorption at 274 nm corresponds to the ⁸S_7/2–⁶I_J transition of Gd³⁺ ions.³⁵ The band gap of Gd₂O₃:Dy³⁺ phosphors is calculated by Kubelka–Munk (K–M) plot and presented in Fig. 5(b). According to the K–M function, the relationship between optical absorption coefficient and band gap energy is given in eqn (4).

(F(R)hν)² = A(hν − E_g) (4)

where h, ν, F(R), A and E_g are the Planck constant, light frequency, K–M function, proportionality constant and energy band gap respectively.³⁶ The band gap is estimated by extrapolating the linear portion of the plot drawn between (F(R) hν)² and hν along the photon energy and the values for 600 °C, 700 °C and 800 °C annealed phosphor are 5.25 eV, 5.18 eV and 5.19 eV respectively.

Fig. 5 (a) Reflectance spectra and (b) band gap plot of Gd₂O₃:Dy³⁺ phosphors annealed at different temperatures.

The PALS measurement is used to evaluate the defect related properties of Gd₂O₃:Dy³⁺ phosphors annealed at different temperatures. The positron lifetime curve is resolved into three components τ₁, τ₂ and τ₃ and the corresponding intensities I₁, I₂ and I₃ are plotted and presented in Fig. 6. The average lifetime of the positron in the phosphors is calculated from eqn (5) which gives the relation between the positron lifetime and intensity of the respective components.

where τ_i and I_i is the lifetime and intensity component of positron (i = 1, 2 and 3).¹⁷ The Table 2 presents the positron lifetime values (τ₁, τ₂ and τ₃), intensities (I₁, I₂ and I₃) and τ_av [τ_av = (τ₁I₁ + τ₂I₂)/(I₁ + I₂)] (τ₃ and I₃ are not considered in the calculation as their contribution is negligibly small and results from the intercrystalline open space region of the particle). The variation of the positron lifetime values of the phosphors confirms the strong interaction between the ions during the annealing process.

Fig. 6 Positron lifetime (a) τ₁ (b) τ₂ (c) τ₃ and intensity (d) I₁ (b) I₂ (c) I₃ of Gd₂O₃ annealed at 800 °C and Gd₂O₃:Dy³⁺ annealed at different temperature.

Table 2 Parameters deduced from positron lifetime spectroscopy

Phosphors	Annealing temperature	τ1 (ns)	τ2 (ns)	τ3 (ns)	τav (ns)	I1 (%)	I2 (%)	I3 (%)
Gd2O3	800 °C	0.220	0.543	5.281	0.292	75.996	22.029	1.975
Gd2O3:Dy^3+	600 °C	0.208	0.522	5.830	0.307	67.052	31.018	1.930
	700 °C	0.206	0.518	4.492	0.301	68.12	29.851	2.029
	800 °C	0.208	0.507	5.002	0.292	70.583	27.461	1.956

---

The first lifetime τ₁ component of undoped Gd₂O₃ phosphor is approximately 0.22 ± 0.0013 ns and the corresponding intensity is approximately 76%. The observed value originate due to the annihilation of positrons at the grains, grain boundaries and their interface.³⁷ Grain boundaries being rich in defects, act as strong trapping centres for positrons and a large fraction of positrons get trapped in the vicinity of grain boundaries rendering a few positrons to annihilate with free electrons inside a grain. In those cases, if the positron lifetime in the grain boundary defects is close to their free annihilation lifetime in defect free regions inside a grain, τ₁ may represent a mixed lifetime comprising of both annihilation at grain, grain boundary and mono vacany in the crystallite as well. The intermediate lifetime τ₂ = 0.54 ± 0.0051 ns, I₂ = 22% is due to the defect species components originating from the annihilation of positron at vacancy clusters located within the grains.³⁸ After thermalization, some positrons escape without annihilating at the grain boundaries and get trapped at the nanovoids (vacancy clusters).

On substituting Dy³⁺ in Gd³⁺ site and annealing at different temperatures affect the positron lifetime to a greater extent. It clearly reveals that the addition of Dy³⁺ in Gd₂O₃ induces more defects at the grain boundary as a result of increased electron density but decreases the τ₁. In Gd₂O₃:Dy³⁺ annealed at different temperatures, τ₁ value remains almost same and I₁ varies gradually from 67% to 71%. During the synthesis, Dy³⁺ ions may get trapped at different locations either individually or forming a bulk clusters. When the phosphor is annealed at different temperatures, the Dy³⁺ ions start migrating and tend to occupy the voids in the matrix and so the electron density around the positron traps increases. As a result, τ₂ and I₂ value decrease with annealing temperatures. Thus, the annealing plays a significant role in reducing the vacancy defects and intensively affect the luminescence properties of Gd₂O₃:Dy³⁺ phosphor (explained in the next session).

The long lifetime τ₃ is due to the formation of positronium by the positron trapped in large voids in the intercrystalline region of the nanoparticles.³⁹ During the annealing process grain growth occurs through the intersection of grain with the neighbouring grain. The interfaces between the nearby grains are the site of very low electron density and it can act as favourable sites for the formation of positronium. In Gd₂O₃:Dy³⁺ annealed at 600 °C, some of the Dy³⁺ ions remains at the surface of the particle and at the intersection of grains. So the ortho-positronium formed in this region has lifetime of 5.83 ± 0.0657 ns and the average crystallite size estimated from XRD is 15 nm. Further on increasing the annealing temperature to 700 °C, grain growth reduces the free volume at the intersection of the grain and thereby increases the crystallite size to 33 nm and decreases the τ₃ (4.492 ± 0.0571 ns) value. At 800 °C, free volume size increases owing to the growth on the boundary line (crystallite size – 44 nm) and hence decrease the electron density and its corresponding τ₃ value increases to 5.002 ± 0.062 ns. The increase in free volume size may be due to the migration of Dy³⁺ ions at the interface. The annihilation of positron through the formation of ortho-positronium intensity (I₃) is only around 2% for all the samples and therefore, the intensity has been excluded in the average lifetime calculation.

From Doppler broadening the S and W parameters covering the low and high momentum region⁴⁰ are deduced for further investigation of defects. Fig. 7 represents the variation of S and W parameters of pure Gd₂O₃ annealed at 800 °C and Gd₂O₃:Dy³⁺ annealed at various temperature. S value dominates for Gd₂O₃:Dy³⁺ annealed at 600 °C due to the presence of large number of Dy³⁺ ions residing on the grain boundary sites. On further increasing the annealing temperature to 700 °C, the S parameter declines and the W-parameter attains a moderate value implying that the Dy³⁺ ions begin to migrate and occupy in the Gd³⁺ sites and thereby, reduce the defect concentration. When the phosphor is annealed at 800 °C, most of the Dy³⁺ ions occupy the Gd³⁺ site of Gd₂O₃ and hence, the value of S and W parameters shift towards the value of pure Gd₂O₃ annealed at 800 °C. The decrease in the S-parameter is certainly due to the reduction of vacancy clusters and the larger fractions of positrons annihilation in the bulk.

Fig. 7 S–W plot of Gd₂O₃ annealed at 800 °C and Gd₂O₃:Dy³⁺ annealed at different temperature.

Fig. 8 shows the excitation spectra of Gd₂O₃:Dy³⁺ phosphors annealed at 600 °C, 700 °C and 800 °C and the phosphors are excited at 572 nm which is the major yellow emission wavelength of Dy³⁺ ions. All the spectra exhibit five distinct peaks at 230 nm, 274 nm, 313 nm, 350 nm and 384 nm with varying intensity. The strong peak at 230 nm denotes the host absorption band (HAB). The small shoulder peak at 256 nm, the intense narrow peak at 274 nm and weak peak at 313 nm correspond to the electron transition from ⁸S_7/2 to ⁶D_J, ⁸S_7/2 to ⁶I_J and ⁸S_7/2 to ⁶P_J lying within the energy levels of Gd³⁺ respectively.⁴¹ The spectral peaks related to the intrinsic f–f transitions in the Dy³⁺ ions are present at 350 nm and 384 nm which is due to the transition from the ground state ⁶H_15/2 to the excited state ⁶P_7/2 and ⁴F_7/2 respectively.¹⁵ From the excitation spectra, it can be observed that annealing the phosphor from 600 °C to 800 °C, the intensity of the peak increases due to the improved crystallinity of the phosphors and the presence of excitation peaks of Gd³⁺ in Dy³⁺ excitation spectra clearly reveals the energy transition from Gd³⁺ to Dy³⁺. The similar energy transfer peaks are observed in Gd(BO₂)₃:Dy³⁺ and in Gd₃Al₅O₁₂:Dy³⁺ phosphor.^15,42

Fig. 8 PL excitation spectra of Gd₂O₃:Dy³⁺ phosphors recorded at 572 nm.

In order to investigate the energy transfer process, the emission spectra of Gd₂O₃:Dy³⁺ phosphors are monitored by exciting Gd³⁺ at 274 nm. In Fig. 9, the emission spectra include a strong yellow emission at 572 nm, blue emission at 486 nm, and a weak red emission at 669 nm (enlarged view is shown as an inset in Fig. 9) which are assigned to the energy transition from the excited state ⁴F_9/2 to three different ground state ⁶H_15/2, ⁶H_13/2 and ⁶H_11/2 respectively. In the electronic states of rare earth ions, magnetic and electric dipole transition occur predominantly where the former is allowed and the later is a forced transitions. Even though the electric dipole transition is forbidden, it occurs when rare earth ion occupies non-inversion symmetry site. In Dy³⁺, the emission peak at 486 nm (⁴F_9/2–⁶H_15/2) is an allowed magnetic dipole transition and is insensitive to the environment.¹⁰ The parity forbidden electric dipole transition at 572 nm (⁴F_9/2–⁶H_13/2) is hypersensitive and strongly affected by the ligand environment. The asymmetric ratio (R) is the integrated intensity ratio of ⁴F_9/2–⁶H_13/2 and ⁴F_9/2–⁶H_15/2 which is used to estimate the symmetry around the Dy³⁺ in the phosphor.⁴³ The asymmetric ratio calculated for Gd₂O₃:Dy³⁺ phosphors annealed at 600 °C, 700 °C and 800 °C are 3.37, 3.21 and 3.26 respectively. All the values are in the order of 3 implying that the Dy³⁺ ions are located at the non-inversion symmetry site in Gd₂O₃. It is very interesting to note that from a single phosphor source it is possible to obtain three different emission colours like blue, yellow and red. The possible energy transfer process from Gd³⁺ to Dy³⁺ is schematically illustrated in the Fig. 10. On exciting the phosphor at 274 nm, ⁶I_J of Gd³⁺ is populated and further the energy is transferred to Dy³⁺ and emission occurs from ⁴F_9/2.

Fig. 9 Emission spectra of Gd₂O₃:Dy³⁺ phosphors recorded at 274 nm.

Fig. 10 Schematic representation of energy transfer process.

Many literatures on phosphor have reported the presence of dominant yellow emission from Dy³⁺ occupying the non-inversion site in YAl₃(BO₃)₄, SrSiO₄ and GdAlO₃.^43–45 Similarly the dominant blue emission is observed for the Dy³⁺ ions occupying the inversion site in GdPO₄, K₂MgP₂O₇ and GdFeO₃ phosphor.^46–48 In case of Gd₂O₃, a sesquioxides with cubic bixbyite structure (Ia3̄) have crystallographically two nonequivalent cationic sites with 6-fold coordination namely C₂ non-inversion symmetry with Gd³⁺ in the centre of a distorted cube with two oxygen vacancies on the face diagonal and S₆ site symmetry with inversion symmetry resulting from two oxygen vacancies lying on one body diagonal.⁴⁹ The photoluminescence properties varies with the site preference. In Gd₂O₃, 75% Gd³⁺ ions occupy the C₂ sites, and 25% Gd³⁺ ions occupy the S₆ sites.^50,51 Stanek et al. have reported the site preference of solute cation (here Dy³⁺) as a function of the ratio between solute and host lattice cation (here Gd³⁺). When the ratio is less than 1, most of the solute cation occupies C₂ site and for value greater than 1, S₆ site is preferred. In the present case, the ionic ratio is 0.972 and so Dy³⁺ prefers C₂ site than S₆.

The luminescence decay curve of Gd₂O₃:Dy³⁺ phosphor is presented in Fig. 11a–c. It is recorded for the dominating wavelength of 572 nm (⁴F_9/2–⁶H_13/2). The curves are fitted with bi-exponential decay presented in eqn (6).

I = A₁e^−t/τ₁ + A₂e^−t/τ₂ (6)

where I is the luminescence intensity at time t, A₁ and A₂ are constants, τ₁ and τ₂ are decay times for the exponential components. The values fitting, are presented inside Fig. 11a–c. It is found that the τ₁ value decreases and τ₂ value increases with temperature. The component τ₁ is associated with the decay of Dy³⁺ available near the surface which is affected by quenching centre. The next parameter τ₂ corresponds to the decay of Dy³⁺ from the interior of the phosphor.^52,53 When the phosphor is annealed, τ₁ value decreases due to migration of Dy³⁺ from the surface into the interior of phosphor and occupies the Gd³⁺ site which leads to the increase in τ₂. It clearly shows that occupancy of Dy³⁺ in Gd³⁺ site, decreases the surface defects.

Fig. 11 Luminescence decay curve of Gd₂O₃:Dy³⁺ annealed at (a) 600 °C (b) 700 °C (c) 800 °C.

The influence of annealing temperature on defect concentration is estimated by comparing and relating the positron average lifetime, the values of S parameter, the yellow emission peak centred at 572 nm and fluorescence lifetime (refer Fig. 12). Upon annealing, the luminescence intensity and fluorescence lifetime τ₂ increases (refer Fig. 12(a)) due to the reduction of lattice defects (refer Fig. 12(b)) which is evident from τ_av and S parameter of PALS.

Fig. 12 Comparison of (a) PL intensity of 572 nm and fluorescence lifetime τ₂ (b) positron τ_av and S parameter as a function of annealing temperature.

Further, to represent the predominant colour emission at each annealing temperature, a CIE (Commission International de I'Eclairage 1931) plot is plotted and depicted in Fig. 13. In this plot, at 600 °C, the emission intensities of blue and yellow are more or less same implying that an equal number of Dy³⁺ ions are substituted in the inversion and non-inversion symmetry site of Gd³⁺. The equal contribution from the blue emission and yellow emission tends to shift the CIE point close to the white region of the chromaticity coordinates. On further annealing the phosphor to 700 °C and 800 °C the blue emission gets suppressed and the yellow emission is enhanced. This is attributed to the non-uniform distribution of Dy³⁺ ions where the large fraction of the ions resides on the non-inversion symmetry site and thus, enhancing and shifting the CIE point towards the yellow region of the chromaticity coordinates. The shift and change in the emission colour are highly reflected in the CIE graph. Thus, by blending an appropriate composition of blue, yellow and red emission, a suitable white colour emission can be achieved in Gd₂O₃:Dy³⁺ phosphor which finds an intensive application in fabricating white LEDs. The CIE coordinates and the quantum yield as a function of annealing temperature are tabulated in the Table 3. The increase in quantum yield with the increase in annealing temperature well supports the fine substitution of Dy³⁺ ions in Gd₂O₃ matrix.

Fig. 13 CIE plot of Gd₂O₃:Dy³⁺ phosphors prepared at different annealing temperatures.

Table 3 CIE coordinated and quantum yield of the phosphors

Annealing temperature	CIE coordinates		Quantum yield (%)
	x	y
600 °C	0.29	0.346	18.52
700 °C	0.347	0.406	21.277
800 °C	0.351	0.408	21.602

---

Conclusions

In summary, Gd₂O₃:Dy³⁺ is successfully synthesized by citrate-based sol–gel process. PALS shows annihilation of positron at various defects in the lattice like grain boundary defects, vacancy, voids and clusters. Positron lifetime studies indicate that the lattice defects created during the synthesis condition gets reduced on annealing. Under Gd³⁺ excitation at 273 nm, Gd₂O₃:Dy³⁺ phosphors exhibit characteristic transition peaks of Dy³⁺ at 486 nm, 572 nm and 669 nm. It clearly indicate a strong energy transfer process from Gd³⁺ to Dy³⁺. The theoretically predicted occupancy of Dy³⁺ ions at C₂ site of Gd³⁺ is also proved experimentally. The luminescence decay value τ₂ increases with temperature due to the reduction of defects and substitution of Dy³⁺ for Gd³⁺. On annealing the phosphors at different temperatures, the emission from Dy³⁺ is enhanced with the decrease in lattice defects. CIE coordinates proved that Gd₂O₃:Dy³⁺ can give a colour very close to the white light. It can also be tuned to emit blue or yellow colour based on the occupancy of Dy³⁺ in Gd³⁺ site.

Acknowledgements

T. Selvalakshmi thank DST INSPIRE Fellowship (2011/297) New Delhi for the financial support. We would like to thank Dr S. Velmathi, Department of Chemistry, National Institute of Technology, Tiruchirappalli – 620015 for DRS and PL measurements and National Centre for Ultra Fast Processes, University of Madras, Chennai for Fluorescence lifetime measurement.

References

1. W.-R. Liu, C.-H. Huang, C.-W. Yeh, J.-C. Tsai, Y.-C. Chiu, Y.-T. Yeh and R.-S. Liu, Inorg. Chem., 2012, 51, 9636–9641.
2. J. A. Dorman, J. H. Choi, G. Kuzmanich and J. P. Chang, J. Phys. Chem. C, 2012, 116, 10333–10340.
3. J. Chen, Q. Meng, P. S. May, M. T. Berry and C. Lin, J. Phys. Chem. C, 2013, 117, 5953–5962.
4. Y. Pan, Q. Zhang, C. Zhao and Z. Jiang, Solid State Commun., 2007, 142, 24–27.
5. L. Wang and Y. Li, Chem. Mater., 2007, 19, 727–734.
6. W. Chen and A. Zhou, J. Phys. Chem. C, 2012, 116, 24748–24751.
7. Z. Xia, J. Zhuang and L. Liao, Inorg. Chem., 2012, 51, 7202–7209.
8. C.-H. Huang, T.-S. Chan, W.-R. Liu, D.-Y. Wang, Y.-C. Chiu, Y.-T. Yeh and T.-M. Chen, J. Mater. Chem., 2012, 22, 20210–20216.
9. E. Pavitra, G. Seeta Rama Raju, Y. H. Ko and J. S. Yu, Phys. Chem. Chem. Phys., 2012, 14, 11296–11307.
10. Z.-W. Zhang, Z.-Y. Moa, S.-T. Song, J.-P. Zhang, L. Liu, W.-G. Zhang and D.-J. Wang, Mater. Lett., 2013, 90, 1–3.
11. X. Yu, L. Zhang, X. Xu, T. Wang, H. Yu, T. Jiang, Q. Jiao, Z. Yang, D. Zhou and J. Qiu, J. Lumin., 2014, 145, 114–118.
12. R.-Y. Yang and H.-L. Lai, J. Lumin., 2014, 145, 49–54.
13. P. V. Do, V. P. Tuyen, V. X. Quang, N. T. Thanh, V. T. T. Ha, H. V. Tuyen, N. M. Khaidukov, J. Marcazzó, Y.-I. Lee and B. T. Huy, Opt. Mater., 2013, 35, 1636–1641.
14. B. P. Kore, N. S. Dhoble, S. P. Lochab and S. J. Dhoble, J. Lumin., 2014, 145, 299–306.
15. X. Zhang, F. Meng, W. Li, S. I. Kim, Y. M. Yu and H. J. Seo, J. Alloys Compd., 2013, 578, 72–76.
16. E. Pavitra, G. S. R. Raju and J. S. Yu, Ceram. Int., 2013, 39, 6319–6324.
17. Y.-H. Lin, M. Li, C.-W. Nan and Z. Zhang, J. Am. Ceram. Soc., 2007, 90, 2991–2994.
18. E. Kuramoto, T. Tsutsumi, K. Ueno, M. Ohmura and Y. Kamimura, Comput. Mater. Sci., 1999, 14, 28–35.
19. Z. Wang, Z. Q. Chen, J. Zhu, S. J. Wang and X. Guo, Radiat. Phys. Chem., 2000, 58, 697–701.
20. R. W. Siegel, Annu. Rev. Mater. Sci., 1980, 10, 393–425.
21. Z. Lu, L. Weng, S. Song, P. Zhang, Q. Hou and X. Ren, J. Sol-Gel Sci. Technol., 2012, 62, 160–169.
22. Y.-H. Lin, M. L. Li, C.-W. Nan and Z. Zhang, J. Am. Ceram. Soc., 2007, 90, 2992–2994.
23. A. V. Thorat, T. Ghoshal, J. D. Holmes, P. M. G. Nambissan and M. A. Morris, Nanoscale, 2014, 6, 608–615.
24. J. Zhang, Z. Zhang, Z. Tang, Y. Tao and X. Long, Chem. Mater., 2002, 14, 3005–3008.
25. H. Saito and T. Hyodo, Radiat. Phys. Chem., 2003, 68, 431–434.
26. C. Suryanarayana and M. Grant Norton, X-ray diffraction-A practical approach, Plenum Press, New York, London, 1998, ch. 3, pp. 207–221.
27. J. Sivasankari, S. Sankar, S. Selvakumar, L. Vimaladevi and R. Kirthiga, Mater. Chem. Phys., 2014, 143, 1528–1535.
28. B. Smith, Infrared Spectral Interpretation-A Systematic Approach, CRC Press LLC, Florida, 1999, ch. 1, pp. 1–29.
29. R. Yogamalar, S. Anita, R. Srinivasan, A. Vinu, K. Ariga and A. C. Bose, J. Nanosci. Nanotechnol., 2009, 9, 5966–5972.
30. S. Lechevallier, P. Hammer, J. M. A. Caiut, S. Mazeres, R. Mauricot, M. Verelst, H. Dexpert, S. J. L. Ribeiro and J. Dexpert-Ghys, Langmuir, 2012, 28, 3962–3971.
31. Y. Jinqiu, C. Lei, H. Huaqiang, Y. Shihong, H. Yunsheng and W. Hao, J. Rare Earths, 2014, 32, 1–4.
32. H. Guo, N. Dong, M. Yin, W. Zhang, L. Lou and S. Xia, J. Phys. Chem. B, 2004, 108, 19205–19209.
33. C. L. Luyer, A. G. -Murillo, E. Bernstein and J. Mugnier, J. Raman Spectrosc., 2004, 34, 234–239.
34. W. A. Pisarski, T. Goryezka, J. Pisarska and W. R. -Romanowski, J. Phys. Chem. B, 2007, 111, 2427–2430.
35. M. L. Pang, J. Lin, J. Fu, R. B. Xing, C. X. Luo and Y. C. Han, Opt. Mater., 2003, 23, 547–558.
36. Z. Jiao, T. Chen, J. Xiong, T. Wang, G. Lu, J. Ye and Y. Bi, Nature, 2013, 3, 1–6.
37. T. E. M. Staab, R. K. -Rehberg and B. Kieback, J. Mater. Sci., 1999, 34, 3833–3851.
38. S. He, C. Li, H. Chen, D. Su, B. Zhang, X. Cao, B. Wang, M. Wei, D. G. Evans and X. Duan, Chem. Mater., 2013, 25, 1040–1046.
39. T. Ghoshal, S. Kar, S. Biswas, S. K. De and P. M. G. Nambissan, J. Phys. Chem. C, 2009, 113, 3419–3425.
40. P. Parz, B. Fuchsbichler, S. Koller, B. Bitschnau, F.-A. Mautner, W. Puff and R. Wurschum, Appl. Phys. Lett., 2013, 102, 151901–151904.
41. M. Jayaimhadri, B. V. Ratnam, K. Jang, H. S. Lee, B. Chen, S.-S. Yi, J.-H. Jeong and L. R. Moorthy, Int. J. Appl. Ceram. Technol., 2011, 8, 709–717.
42. J. Li, J.-G. Li, S. Liu, X. Li, X. Sun and Y. Sakka, J. Mater. Chem. C, 2013, 1, 7614–7622.
43. G. V. L. Reddy, L. R. Moorthy, B. C. Jamalaiah and T. Sasikala, Ceram. Int., 2010, 39, 2675–2682.
44. S. K. Gupta, M. Kumar, V. Natarajan and S. V. Godbole, Opt. Mater., 2013, 35, 2320–2328.
45. G. S. R. Raju, J. Y. Park, H. C. Jung, H. K. Yang, B. K. Moon, J. H. Jeong and J. H. Kim, Opt. Mater., 2009, 31, 1210–1214.
46. A. I. Becerro, S. R. -Liviano, A. J. F. -Carrion and M. Ocana, Cryst. Growth Des., 2013, 13, 526–535.
47. R. L. Kohalea and S. J. Dhoble, J. Lumin., 2013, 138, 153–156.
48. Y. Zhang, A. Zheng, X. Yang, H. He, Y. Fan and C. Yao, CrystEngComm, 2014, 14, 8432–8439.
49. C. R. Stanek, K. J. McClellan, B. P. Uberuaga and K. E. Sickafus, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 134101–134107.
50. E. Antic-Findancev, J. Hosla and M. Lastusaari, J. Alloys Compd., 2002, 341, 82–86.
51. E. Antic-Fidancev, J. Hölsä and M. Lastusaari, J. Phys.: Condens. Matter, 2002, 15, 863–876.
52. A. I. Becerro, S. R. -Liviano, A. J. F. -Carrion and M. Ocana, Cryst. Growth Des., 2013, 13, 526–535.
53. A. K. Parchur, A. I. Prasad, S. B. Rai, R. Tewari, R. K. Sahu, G. S. Okram, R. A. Singh and R. S. Ningthoujam, AIP Adv., 2012, 2, 032119.

---