Structural, optical and special spectral changes of Dy³⁺ emissions in orthovanadates

Abstract

This paper reports on the structural, optical and photometric characterization of yttrium gadolinium orthovanadates (Y_1−xGd_xVO₄) doped with Dy³⁺ for white emission in solid state lighting. A series of orthovanadates were prepared through a low temperature co-precipitation method and further post annealed. The as synthesized phosphor particles revealed a single phase tetragonal structure with space group I4₁/amd(141). The infrared spectra confirmed the presence of characteristic vibrational bands of orthovanadates. The microscopic images showed elongated particles after annealing and the particle sizes were estimated in the range of 10–50 nm. The band gap of the prepared phosphors, calculated from the corresponding diffuse reflectance spectra was observed to be 3.75 eV and 3.57 eV for YVO₄ and GdVO₄ respectively. Y_1−xGd_xVO₄:Dy³⁺ phosphors, illuminated with ultraviolet light exhibited characteristic blue and yellow luminescence corresponding to ⁴F_9/2 → ⁶H_15/2, ⁴F_9/2 → ⁶H_13/2 transitions of Dy³⁺ ion. The emission spectra showed the variation in the intensity ratio (Y/B) with Gd³⁺ ion variation. Furthermore the thermal quenching property, decay analysis and photometric characterizations were also studied in detail and the results indicated the suitability of these phosphors in solid state lighting.

---

1. Introduction

Phosphor converted light emitting diodes related to solid state lighting is the next generation lighting technology and it has become the topic of intensive research in the present era. Achieving white light emission from a single phosphor component is the main target of researchers in this area. Relatively less complex compositions with high brightness and good color purity and low power consumptions are the main criteria for the desired phosphors suitable for white light emitting devices.^1,2 Therefore, to achieve phosphors with the mentioned characteristics, there is a growing trend in developing a single phase phosphor with high efficiency which can emit white light with good colour purity. Low dimensional materials are helpful in achieving good chemical and physical performances for promising applications.^3,4 To date a significant amount of research effort has been devoted on developing novel luminescent materials.^5,6

Metal oxides possessing alkaline, transition or lanthanide exhibit excellent photoluminescence when doped with lanthanide ions (e.g. Eu³⁺, Tb³⁺, Dy³⁺), especially lanthanide metal oxides due to the equivalent chemical properties of dopant and replaced ions.^7,8 Among the various orthovanadates YVO₄ and GdVO₄ have received intensive attention due to their efficient luminescence and diverse applications. Both YVO₄ and GdVO₄ have same crystal structures, and about same melting points owing to which the mixed vanadate should not deviates the optical properties. Park et al. examined significant enhancement in PL intensity in addition of Gd³⁺ in YVO₄:Eu³⁺.⁹ Compare to YVO₄, GdVO₄ is considered as a better host and lasing material owing to the superior qualities of gadolinium such as long luminescence lifetime, narrow emission bandwidth and high resistance to photobleaching. The lanthanide doped inorganic nanoparticles have found versatile applications in bio imaging. The presence of unpaired electrons in Gd³⁺ ion makes it paramagnetic in nature which has been used as an MRI contrast agent.^10,11

A detail literature survey helped us to conclude that Y_xGd_1−xVO₄ (doped with Dy³⁺) as a mixed vanadate phosphor has not been studied as a downconverting phosphor and the effect of the ratio/concentration of Y³⁺ and Gd³⁺ on PL emission and excitation spectra of doped lanthanide. Chen et al. have studied the luminescence of Eu³⁺ and Bi³⁺ in different composition of Y, Gd orthovanadate i.e. (Y,Gd)VO₄.¹² In view of higher atomic weight of Gd than that of Y the thermal stability of dopant luminescence is expected to be more in GdVO₄ rather than YVO₄. Due to the ⁴F_9/2 → ⁶H_13/2 and ⁴F_9/2 → ⁶H_15/2 transitions of Dy³⁺ ion in the blue and yellow regions of the electromagnetic spectrum, it has become an interesting activator, because on the selection of proper host and concentration of Dy³⁺ ions white light emission can be achieved.^13,14 In a mixed crystal host the local crystal field experienced by Dy³⁺ ions may vary. Therefore, doping of Dy³⁺ ions in mixed vanadates and the relevant studies of crystal field effect on splitting of energy levels and their transitions can bring a new understanding of the spectral behaviour of lanthanides.

In this study, we report the photoluminescence properties, decay mechanisms, color chromaticity and the possible applications of Y_1−xGd_xVO₄ doped with Dy³⁺ for UV excited solid state lighting. The structural properties were measured via X-ray diffraction and scanning electron microscopy as well as transmission electron microscopy. The absorption capability was confirmed via the UV/Vis reflectance measurements. The photoluminescence results indicate the proposed phosphor exhibited good emission intensity. These results demonstrated that the present system is a promising near white-emitting phosphor.

2. Experimental

Y_1−xGd_xVO₄ doped with yDy³⁺ has been synthesized using co-precipitation method. Dy³⁺ was taken as substitutional dopant in place of Y³⁺ as well as Gd³⁺. The following compositions were prepared: YDy_0.002VO₄ and 0.2 mol% Dy³⁺ doped Y_1−xGd_xVO₄ (x = 0.05, 0.10, 0.25, 0.50, 0.75, and 1.0). In a typical experiment, stoichiometric amount of precursor yttrium oxide (Y₂O₃), gadolinium oxide (Gd₂O₃), europium oxide (Eu₂O₃) and ammonium meta vanadate (NH₄VO₃) of high purity (99.99%) were taken. Oxide precursors were dissolved in a fixed amount of nitric acid followed by heating to form their nitrates separately.

NH₄VO₃ is dissolved in the solution of 0.05 M per L of NaOH through vigorous stirring and maintained the pH at 10. In the prepared NH₄VO₃ solution nitrates were added dropwise with a continuous rate (1 drop/3 s), at temperature 60 °C followed by stirring. The pH of the obtained solution was maintained at 9 through 1 h of stirring. Slowly white precipitate was formed which was filtered and dried at 80 °C for 6 h. After that all the samples were annealed at 500 °C for 2 h to remove the unwanted carboxyl groups which act as luminescent quenchers. To see the effect of further heat treatment the prepared powders were annealed at different temperatures. The former reaction can be written as

NH₄VO₃ + (1 − x)Y(NO₃)₃ + xDy(NO₃)₃ → Dy_x:Y_1−xVO₄ + NH₄⁺ + 9NO₃⁻

To identify the phase structure and determine the mean size of the micro-crystals X-ray diffraction (XRD) analysis was carried out with a powder diffractometer (Bruker D8 FOCUS) using Cu Kα (0.15406 nm) radiation. The microstructures of the samples were studied using field emission scanning electron microscope (FESEM Supra55) equipped with the energy dispersive X-ray (EDX) spectroscopy system. High resolution transmission electron microscopy (HRTEM) of prepared samples was studied using JEOL JEM-2100 HRTEM measuring instrument to examine the particle size and surface morphology. The excitation and emission spectra were recorded on Hitachi FL-2500 fluorescence spectrophotometer equipped with a xenon lamp as an excitation source. For the comparison of emission and excitation intensities of different samples, all the measurements were done at the same instrumental parameters such as same excitation wavelength and same power, same excitation and emission slits. All the characterizations were carried out at room temperature.

3. Results and discussion

3.1 Crystal structure and phase identification

XRD patterns of 0.2% Dy³⁺ doped YVO₄, GdVO₄ is shown in S1 (ESI data†). It is evident from the Fig. S1† that all the samples are crystalline after annealing at 500 °C for 2 h with tetragonal structure. All diffraction peaks matched well with the JCPDS card no. 72-0274 (a = 7.1 Å, c = 6.27 Å and V = 316.07 ų) for Dy³⁺ doped YVO₄. Similarly diffraction peaks are assigned with card no. 72 0277 (a = 7.19, c = 6.33 and V = 327.23 ų) for GdVO₄. The diffraction data of other three samples Y_0.75Gd_0.25VO₄, Y_0.50Gd_0.50VO₄ and Y_0.25Gd_0.75VO₄ were found to be matched well with JCPDS card no. 85 2318 (a = 7.136, c = 6.3 and V = 320.81 ų), 85-2317 (a = 7.143, c = 6.31, V = 321.95 ų) and 85-2316 (a = 7.164, c = 6.315 and V = 324.1 ų) respectively (Fig. S2†). As synthesized 0.2% Dy³⁺ doped Y_0.25Gd_0.75VO₄ and annealed at different temperatures along with the JCPDS pattern is shown in Fig. 1. No any other diffraction line related to residue or dopant is appeared in all samples. To calculate the crystallite size of the prepared samples Debye–Scherrer formula was used.¹⁵where λ is the wavelength of X-ray used, β is the FWHM of diffraction line and θ is the Bragg's angle.

Fig. 1 (a) XRD pattern of as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ synthesized at various temperatures. Standard JCPDS pattern is also given for reference. (b) Magnified XRD patterns of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ synthesized at various temperatures.

Average crystallite size obtained from the calculation is found to be in the range of 2–28 nm. The result was supported by the calculation done by utilizing Williamson–Hall relation:¹⁶

and plotting 0 straight line graph of β cos θ/λ versus sin θ/λ, inverse of intercept on y-axis of which gave the average crystallite size. The XRD patterns of Y_0.25Gd_0.75VO₄ annealed at different temperatures resulted in narrowing the diffraction lines. The crystallite size is increased with increasing annealing temperatures from 500 °C to 800 °C which supports the earlier report.¹⁷ For the as synthesized sample crystallite size was in the range of 2–7 nm which became 10–48 nm for annealed samples. Above 600 °C, there is no considerable change in the crystallinity of the sample (Fig. 1). Williamson–Hall plot for Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ is shown in Fig. 2. Table 1 lists the average crystallite size calculated from Debye–Scherrer relation and Williamson–Hall relation. A shifting in XRD lines towards lower angle is clearly visible in the magnified XRD patterns of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ synthesized at various temperatures as shown in Fig. 1(b). Such peak shift is related to the lattice expansion after sintering.

Fig. 2 Williamson–Hall plot of Y_0.25Gd_0.75VO₄ annealed at (a) 500 °C (b) 600 °C (c) 700 °C and (d) 800 °C.

Table 1 Average crystallite size calculated from XRD data

Y0.25Gd0.75VO4:0.2% Dy^3+	Crystallite size calculation (in nm)
	Debye–Scherrer	Williamson–Hall
As Syn	2–7	6
500 °C	10–30	27
600 °C	18–37	40
700 °C	18–42	46
800 °C	18–46	48

---

3.2 IR study

Fig. 3 shows the FTIR spectra of as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ along with annealed at different temperatures. From the spectra for as synthesized sample, a strong absorption peak at 818 cm⁻¹ and a weak one at 437 cm⁻¹ have appeared, which are attributed to the absorption of V–O stretching and Y(Gd or Dy)–O bonds, respectively. A strong absorption peak at 3450 cm⁻¹ and a weak one at 1642 cm⁻¹ can be assigned into the symmetric stretching vibration and bending vibration of H–O–H (H₂O molecules), respectively.^18,19 An absorption peak at 1396 cm⁻¹ can be assigned into asymmetric stretching vibration of –NO₃⁻ which comes during the synthesis process. By increasing the annealing temperatures the bands related to H–O–H and –NO₃⁻ at 3450 cm⁻¹, 1642 cm⁻¹ and 1396 cm⁻¹ became diminished as their presence was due to synthesis condition. Samples synthesized through wet chemical route can retain these types of bands along with them until high temperature is applied. For the sample annealed at 600 °C and other higher temperature the H–O–H band situated at 3450 cm⁻¹ is completely removed. Annealing at 800 °C made the sample free from all luminescent quencher bands and V–O stretching band is slightly shifted towards lower wavenumber due to the lattice expansion with higher temperature. Thus, the single phase Y_0.25Gd_0.50VO₄:0.2% Dy³⁺ phosphor formation is also confirmed by FT-IR studies.

Fig. 3 FTIR spectra of as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ and annealed at different temperatures.

3.3 Surface morphology

As all the samples were prepared at same synthesis condition so there was no obvious change in surface morphology with the variation in Y and Gd concentration was observed. Although annealing at different temperatures brought some morphological changes. In Fig. 4, the field emission scanning electron microscope (FESEM) images of the sample Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at various temperatures are shown. The as synthesized sample is found to be in agglomerated form and there is no big change after annealing at 500 °C for 2 h was observed. On varying the annealing temperature the morphological changes occurred. Sample annealed at 500 °C was spherical in shape while the samples annealed at 600 °C resulted in rod shape formation, the length of which increases with increasing annealing at 700 °C and 800 °C. Whereas annealing at 800 °C brought an agglomeration.

Fig. 4 FESEM images of as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ with EDS (a) as synthesized and annealed at (b) 500 °C (c) 600 °C (d) 700 °C (e) 800 °C. Inset of (c): EDS pattern of the sample annealed at 600 °C.

The possible reason behind this is the thermodynamical instability of nanomaterials due to the presence of a large number of possible interface boundaries. There is a strong tendency for nanocrystalline materials to convert in conventional coarser grain materials with fewer interfaces. Wang et al.²⁰ pointed that the crystallites growth is a result of grain boundary motion, which is driven by two processes: (i) grain boundary diffusion and (ii) grain boundary migration. Here also the possibility of grain boundary formation with annealing increases which led to the rod shape formation from spherical nanoparticles. In Fig. 4 the corresponding scales has been given for clear demonstration. The confirmation of elemental composition done by EDS study is also shown for the sample annealed at 600 °C for 2 h as the inset of Fig. 4(c).

To determine the particle size and the crystallinity of the as synthesized sample more precisely, transmission electron microscope (TEM) and selected area electron diffraction (SAED) of the as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at 500 °C are carried out, as shown in Fig. 5. The TEM images indicate nanoparticles of sizes around 8–35 nm in diameter. This finding is consistent with the crystallite size evaluated from the powder X-ray diffraction (XRD) measurements. In the high-resolution transmission electron microscopy (HRTEM) images of the present sample (Fig. 5(c)), the lattice fringes on the individual nanoparticle are clearly distinguishable, indicating that the prepared nanoparticles are highly crystalline. The distance between the lattice fringes of the undoped sample was measured to be 0.42 nm, which corresponds to the d-spacing of lattice plane in the tetragonal YVO₄. In particular, similar values of the crystalline domain size and microscopically estimated average particle size of the nanoparticles imply that each particle consists of a single crystallite. The SAED pattern (inset of Fig. 5(a)) shows features typical to polycrystalline powders consist of small particles. The particle size distribution shows most of the particles in the range of 10–30 nm (Fig. 5(d)).

Fig. 5 (a) TEM image (b) magnified TEM image, (c) HRTEM image and (d) particle size distribution for the as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at 500 °C. Inset of Fig. 1(a): corresponding SAED pattern.

3.4 Bandgap calculation

The DR spectra of Dy³⁺ doped YVO₄, GdVO₄ and Y_1−xGd_xVO₄ (x = 0.25, 0.5, 0.75) phosphors were measured against spectralon, a commercially available material made from polytetrafluoroethylene (PTFE), which is used as a reference sample during the diffuse reflectance process. In the diffuse reflectance spectra, a sharp band at 310 nm was observed for all samples as shown in Fig. 6. The band at 310 nm was due to the band gap of the prepared phosphors. In the case of Y_0.25Gd_0.75VO₄ phosphor a band edge was observed at 310 nm. This is due to the charge transfer from V⁵⁺ to O²⁻.²¹ No other weak bands beyond 350 nm were observed for Dy³⁺ doped samples. This can be due to the low concentration of dysprosium ions. The Kubelka–Munk theory was used to calculate the band gap of the synthesized phosphors using the DR spectra. In the DR spectra, the ratio of the light scattered from a thick layer of the sample and an ideal non-absorbing reference sample is measured as a function of the wavelength λ, R_∞ = R_sample/R_reference. The relationship between the diffuse reflectance of the sample (R_∞), the absorption coefficient (K) and the scattering coefficient (S) is given by the Kubelka–Munk function F(R_∞):²²

Fig. 6 Diffuse reflectance plot of as synthesized 0.2% Dy³⁺ doped samples annealed at 500 °C. Inset: Kubelka–Munk plot.

The band gap E_g and linear absorption coefficient α of a material are related through the well-known Tauc relation:²³

(αhν)² = A(hν − E_g)ⁿ (4)

where ν is the photon energy and A is a proportionality constant. When the material scatters in a perfectly diffuse manner, the absorption coefficient K becomes equal to 2α. Considering the scattering coefficient S, a constant with respect to the wavelength, and using above two equations, the following expression can be written as:

[F(R_∞)hν]² = B(hν − E_g)ⁿ (5)

The value of n is 1 for direct allowed transitions, 2 for nonmetallic materials, 3 for direct forbidden transitions, 4 for indirect allowed transitions and 6 for indirect forbidden transitions.²⁴

GdVO₄ and YVO₄ are direct band gap materials so for them the value of n chosen to be 1. But there is no previous reports on the band gaps of Y_1−xGd_xVO₄ materials so all the values of n were checked. Among the plots of [F(R_∞)hν]², [F(R_∞)hν], [F(R_∞)hν]^2/3, [F(R_∞)hν]^1/2 and [F(R_∞)hν]^1/3 as a function of the photon energy hν, the best fitting was found with n = 1 in above equation, i.e., [F(R_∞)hν]² as a function of hν indicating that the band transitions that occurred were direct in nature, as suggested by Tauc et al.²³ From the plot of [F(R_∞)hν]² versus hν, the value of E_g was obtained by extrapolating the linear fitted regions to [F(R_∞)hν]² = 0. The band gap calculated from the DR spectra using the K–M function F(R_∞) was found to be around 3.75 eV and 3.57 eV for YVO₄ and GdVO₄ respectively. The estimated values of band gap for Y_0.75Gd_0.25VO₄, Y_0.5Gd_0.5VO₄, Y_0.25Gd_0.75VO₄ samples were 3.57, 3.44 and 3.54 eV, respectively, which are lowered than that of YVO₄. The reason behind lowering of band gap with the addition of Gd ion is the formation of meta-stable states in between the valence band and conduction band.²⁵

3.5 Photoluminescence study

Fig. 7 shows the excitation spectra of Y_1−xGd_xVO₄:0–0.2% Dy³⁺ (x = 0, 0.25, 0.50, 0.75 and 1.00) phosphor (prepared at ph 9) by monitoring the emission wavelength at 575 nm which corresponds to electric dipole transition of Dy³⁺ ion.²⁶ The excitation spectra consist of a strong peak situated at 310 nm which is referred as V–O charge transfer band of VO₄³⁻ group, which occurs in VO₄³⁻ through electron transfer from oxygen 2p state to the empty d states of central vanadium ion. Apart from this broad peak there are some peaks of relatively low intensities were also observed above 350 nm. These small peaks are related to f–f transitions of Dy³⁺ ions at wavelength 366, 389, 427, 451 and 474 nm respectively which corresponds to transition from ground level of Dy³⁺ to other higher level which are ⁶H_15/2 → ⁶P_5/2, ⁶H_15/2 → ⁴I_13/2 + ⁴F_7/2, ⁶H_15/2 → ⁴G_11/2, ⁶H_15/2 → ⁴I_15/2 and ⁶H_15/2 → ⁴F_9/2 transitions.²⁷ It is clearly viewed in the spectra that V–O absorption band is much dominating over f–f absorption of Dy³⁺ ion. The reason behind the strong absorption of VO₄³⁻ is its high absorption co-efficient i.e. 10³ to 10⁵ cm⁻¹ compared to Dy³⁺ absorption cross section (0.1–10 cm⁻¹) which is primarily forbidden by selection rules.²⁸ YVO₄ has the absorption in 300–320 nm range and emission in the region of 350 to 510 nm wavelengths, which almost overlaps the f–f absorption of Dy³⁺ ion shown in S4 (ESI data†). The overlapping of VO₄³⁻ emission and Dy³⁺ absorption leads to efficient energy transfer from vanadate group to Dy³⁺ ion. As the excited photon from VO₄³⁻ emission are absorbed by activator (Dy³⁺) because the excitation energy levels of Dy³⁺ ion lie on the emission levels of VO₄³⁻ in the range of 370–510 nm. This clearly indicates that the incident energy was successfully absorbed by VO₄³⁻ group and it will lead to sufficient energy transfer to Ln³⁺ ion hence vanadate group acts as a sensitizer.

Fig. 7 Excitation spectra of as synthesized 0.2% Dy³⁺ doped samples annealed at 500 °C. Magnified spectrum of f–f transition of Dy³⁺ ion.

Along with this all lanthanide ions need a perfect host which allow crystal field to split the energy levels of the doped ions. Orthovanadate provides D_2d symmetry for Ln³⁺ ions. This crystal symmetry gives an asymmetric environment to doped lanthanide ion. Hence an expectation of domination of electric dipole transition over magnetic dipole increases.²⁹ The emission spectrum of Dy³⁺ doped phosphors were recorded at an excitation wavelength of 310 nm.

Fig. 8 shows emission spectra of 0.2% Dy³⁺ doped YVO₄ phosphor. It consists of two main distinct peaks at 485 nm and 575 nm which correspond to ⁴F_9/2 → ⁶H_15/2 and ⁴F_9/2 → ⁶H_13/2 transitions respectively and owing to non-radiative energy migration from VO₄³⁻ to activator sites the emission from VO₄³⁻ is very weak in the region of 350 to 460 nm. The 310 nm radiation excites the Dy³⁺ ions to ⁶P_7/2 level and then it quickly come back to the ⁴F_9/2 via non radiative transition. The strong yellow emission band centered at 575 nm is related to the hyper sensitive electric dipole transition ⁴F_9/2 → ⁶H_13/2 of Dy³⁺ ion. Another blue emission band situated at 485 nm corresponds to ⁴F_9/2 → ⁶H_15/2, which is comparatively less sensitive to the host.³⁰ It is clear from the spectrum that ⁴F_9/2 → ⁶H_13/2 transition at 575 nm is more prominent than the ⁴F_9/2 → ⁶H_15/2 transition at 485 nm. There was no any other transitions related to Dy³⁺ or VO₄³⁻ was observed in the emission spectra.

Fig. 8 PL emission spectrum of the as synthesized YVO₄:0.2% Dy³⁺ annealed at 500 °C.

3.5.1 Effect of Gd³⁺ concentration. To see the effect of Gd³⁺ contents on spectral behavior of Dy³⁺ emission mixed vanadates are prepared i.e. Y_1−xGd_xVO₄:0.2% Dy³⁺ (x = 0.25, 0.50, 0.75 & 1.0). The concentration of Gd³⁺ has been varied from 25 mol% to 100 mol%. Fundamentally, the emission properties of phosphor materials strongly depend on the synthesis route, size of the particles, host environment and concentration of activator ions.^31,32 As addition of Gd not only changed the crystal structures but also affected the crystal field generated by the different crystal environments. The intensities of ⁴F_9/2 → ⁶H_15/2 and ⁴F_9/2 → ⁶H_13/2 transitions are varied with varying crystal symmetry. The enhancement in degree of emission after Gd addition exists because of host lattice environment. Among all the varying concentration of gadolinium Y_0.25Gd_0.75VO₄ doped with 0.2% Dy³⁺ exhibited maximum PL emission as shown in Fig. 9. Intensity of Dy³⁺ emission depends strongly upon the crystal environment and among all the presently studied orthovanadates Y_0.25Gd_0.75VO₄ has provided the best crystal environment for Dy³⁺ doping. In mixed orthovanadates, Dy³⁺ ions can replace Y³⁺ as well as Gd³⁺ ions.

Fig. 9 PL Emission spectra of the as synthesized 0.2% Dy³⁺ doped Y_1−xGd_xVO₄ annealed at 500 °C.

In the excitation spectra (Fig. 7) it can be observed that the charge transfer band (CTB) became much dominating compare to the doped YVO₄ phosphor; this leads to the fact that Gd³⁺ ion addition contribute to the charge transfer phenomenon. Gd³⁺ has a transition at 250 nm which is related to 8S–6D transition of Gd³⁺ ion which brought an enhancement in the broad band excitation spectra.³³ The wide excitation band below 375 nm has two possible origins, that is, the charge-transfer band (CTB) between Dy³⁺ and O²⁻ or CTB between V⁵⁺ and O²⁻. Usually the charge transfer of V⁵⁺–O²⁻ in VO₄³⁻ occurs much more easily than that of Dy³⁺–O²⁻ due to the large differences in charges and ionic radii between V⁵⁺ (r = 0.355 Å) and Dy³⁺ (r = 0.91 Å) ion.⁶ From the viewpoint of molecular orbital theory, the excitation such as 310 nm in Y_1−xGd_xVO₄:Dy³⁺ could corresponds to electric or magnetic dipole allowed transitions from the ¹A₂ (¹T₁) ground state to ¹E (¹T₂), ¹A₁ (¹E), and ¹B₁ (¹E) excited states of VO₄³⁻ ions, respectively. The band at around 250 nm assigned to ¹A₁ → ¹T₂ (t₁ → e) (symmetry allowed) and the band at 310 nm assigned to the ¹A₁ → ¹T₁ (t₁ → e) (symmetry forbidden) transition of VO₄³⁻ ion.^6,34 In Dy³⁺-doped YVO₄, it has been proposed that the contribution of Dy³⁺–O²⁻ charge transfer overlapped with the excitation band (220–350 nm) ascribed to a charge transfer from the oxygen ligands (O²⁻) to the central vanadium atom (V⁵⁺) within the VO₄³⁻ group.³⁵ As in all the studied samples the concentration of Dy³⁺ was kept fixed i.e. 0.2 mol% so the chances of enhancement in CTB related to Dy³⁺–O²⁻ charge transfer is quite less after Gd³⁺ addition. However this enhancement could be due to charge transfer from V⁵⁺ to O²⁻ ion.

As ⁴F_9/2 → ⁶H_13/2 transition at 576 nm is hyper sensitive, its intensity can vary by orders of magnitude depending on the local environment. Hence ⁴F_9/2 → ⁶H_13/2/⁴F_9/2 → ⁶H_15/2 (Y/B) emission ratio called as asymmetric ratio (A₂₁) can be used as a probe to measure the site symmetry or structural changes in the vicinity of the Dy³⁺ ion and covalency of the bonds with the oxygen ions.³⁶ Asymmetry ratio is calculated by using the relation:

where I₁ and I₂ represent the respective integrated intensities of ⁴F_9/2 → ⁶H_15/2 and ⁴F_9/2 → ⁶H_13/2 transitions of Dy³⁺, respectively.

In the previous reports on Y/B ratio the variation was explained for varying concentration of Dy³⁺.³⁷ Whereas the local environment and asymmetric ratio can be tailored by the variation of growth ambience and modification in the host lattice which can influence the performance of luminescent material. For Dy³⁺ doped YVO₄, the obtained Y/B ratio was 0.77 which increased upto 0.99 for Y_0.50Gd_0.50VO₄ and again decreases to 0.82. It is worth to mention that if the site occupied by the Dy³⁺ ions has a high symmetry then the Y/B ratio should be low. Whereas lower symmetry results to the high values of Y/B.³⁸ In our studies, the change in ratio is not drastic with Gd³⁺ variation. It is known that both YVO₄ and GdVO₄ host lattices offer one symmetric site with point group D_2d. Therefore, low values of Y/B were expected since Dy³⁺ ions occupying a relatively high symmetry site. Additionally, one can note that intensities of the both transitions are comparable.³⁹ This could be directly related to the structural distortion of the site occupied by Dy³⁺ and more importantly to the strong anisotropy of the host crystals. In the mixed vanadates, Dy³⁺ ions occupy both the sites provided by Y³⁺ and Gd³⁺ which affects the symmetry provided by the host. Therefore, an enhancement in asymmetry ratio was observed. The full width half maxima for all the samples were found to be greater for the peak situated at 485 nm than the transition assembled at 575 nm. Although the band at 575 nm is more intense than that of 485 nm but the broadening of the band reduced the asymmetry ratio. This also indicates that the Y_0.50Gd_0.50VO₄ ligand is less symmetric one. Table 2 lists the relative peak width and asymmetry ratios.

Table 2 Asymmetry ratio and effective peak widths

Y1−xGdxVO4:0.2% Dy^3+ x =	Y/B ratio	Peak width (in nm)
		^4F9/2 → ^6H13/2	^4F9/2 → ^6H15/2
0	0.77	11.20	7.38
0.25	0.84	11.12	7.62
0.50	0.99	10.34	7.41
0.75	0.82	10.49	7.48
1.00	0.82	11.79	7.59

---

3.5.2 Effect of annealing. Effect of heat treatment on the PL of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ was studied by varying the annealing temperature from 500 to 800 °C. As seen from Fig. 10, the sample annealed at 600 °C exhibited the best PL intensity. The as synthesized sample showed no PL owing to the less crystallinity and high defects. However, when the samples were allowed for heat treatment for two hrs at 500 °C and 600 °C respectively, the crystallinity increased drastically which was previously confirmed by XRD and FESEM studies (Fig. 1 and 4, respectively). As synthesized samples were amorphous in nature, shown in Fig. 1 and the presence of surface dangling bonds such as carboxyl and hydroxyl groups which act as luminescent quenchers led to no PL emission. After annealing, these groups were removed and the emission of Dy³⁺ ions was observed. Fig. S4 (ESI data†) shows the excitation spectrum of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at 600 °C for 2 h when monitored at the emission wavelength of 576 nm. It consists of a broad pick around 310 nm which is related to V–O charge transfer. The Dy–O charge transfer band could also be observed as a small hump in the excitation spectrum around 280 nm. Post annealing can bring these types of changes in the prepared phosphor (a) increases the crystallinity, (b) morphological change and (c) removal of surface dangling bonds (–CH, –OH) (d) non-radiative defect creation.⁴⁰

Fig. 10 PL Emission spectra of as synthesized Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at different temperatures.

All the mentioned points have direct connection to luminescence process. Decrease in PL intensity of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed above 600 °C was observed. Yoshiki Iso et al. have reported a detailed mechanism behind the decrement in PL intensity due to the thermal treatment for YVO₄:Eu³⁺, Bi³⁺ phosphor.⁴¹ Herein, the decrease in PL intensity above 600 °C could be due to the fact that during high-temperature synthesis, the excess thermal energy deteriorated the crystallinity which caused in a distortion of the crystal field rather than enhancing the atomic arrangement for the recrystallization. Furthermore, the change in morphology could also be the reason to decrease the PL intensity. At 600 °C, the sample exhibited rod shaped morphology the length of which increases due to the heat treatment above 600 °C (Fig. 4). Therefore, the sample annealed at 600 °C has a higher specific surface area than that annealed above 600 °C, which can be easily distinguished from the SEM images (Fig. 4). Generally, if the specific surface area of any sample decrease, the luminescent centers on the crystal surface also decrease. However, the sample annealed at 500 °C has larger surface area than that annealed at 600 °C. Such larger surface area exposed luminescent centers to interact with their surrounding defects such as the –CH and –OH groups, owing to which the PL of 500 °C annealed sample is observed to be lesser than that annealed at 600 °C, and finally the emission was partly quenched through the multiphonon relaxation process.⁴²

3.5.3 Decay kinetics. Fig. 11 shows the decay curve of Dy³⁺ doped Y_1−xGd_xVO₄ (x = 0, 0.75, 1) samples recorded for the 575 nm emission of Dy³⁺ ions (λ_exi = 310 nm). The resultant decay curves were fitted via monoexponential and bi-exponential function. However, the best fit was obtained for biexponential decay:

I(t) = A₁e^−t/τ₁ + A₂e^−t/τ₂ (6)

where I(t) is the luminescence intensity at time t and A₁ and A₂ are the luminescence intensity at time τ₁ and τ₂ respectively and τ₁ and τ₂ are the decay times for the exponential components, respectively. The effective lifetime can be evaluated from the equation:

Fig. 11 Luminescence decay curve of 0.2% Dy³⁺ doped (a) YVO₄ (b) Y_0.25Gd_0.75VO₄ (c) GdVO₄.

The bi-exponential decay occurred in the studied samples due to the non-uniform distribution of Dy³⁺ ions in the host matrix which led to the variation in local concentration.⁴³ Transfer of excitation energy from the sensitizer to the activator also resulted to multiexponential decay. The effective lifetime calculated for 2 mol% Dy³⁺ doped YVO₄, Y_0.25Gd_0.75VO₄ and GdVO₄ are 0.318 ms, 0.410 ms and 0.388 ms, respectively. The effective lifetime of Dy³⁺ ion corresponding to the blue emission (485 nm) of Dy³⁺ ions in above samples were also calculated and the fitted graph is shown in S5 (ESI data†) and its values were 0.318 ms, 0.429 ms and 0.426 ms, respectively. The resultant values of emission life times showed very less dependence on the host lattice environment.

3.5.4 Photometric characterizations. Fig. 12 illustrates the CIE (International Commission on Illuminations) chromaticity diagram of Y_1−xGd_xVO₄:0.2% Dy³⁺ phosphor for different Y³⁺/Gd³⁺ mass ratios. The CIE parameters such as colour coordinates (x, y) and the colour co-related temperatures were calculated by using the colour calculator software to characterize the colour emission of prepared phosphors. The colour of any light source can be described just by three variables called colour matching functions and are expressed by dimensional quantities x̄(λ), ȳ(λ) and z̄(λ).⁴⁴ For a given power-spectral density P(λ), the degree of stimulation required to match the colour of P(λ) is given by three equations based on three tristimulus values X, Y and Z.

Fig. 12 CIE diagram of the as synthesized 0.2% Dy³⁺ doped (a) YVO₄ (b) Y_0.75Gd_0.25VO₄ (c) Y_0.5Gd_0.5VO₄ (d) Y_0.25Gd_0.75VO₄ (e) GdVO₄ annealed at 500 °C.

The colour co-ordinates x and y were calculated by using the above values through following relations

The colour co-ordinates for Y_1−xGd_xVO₄:0.2% Dy³⁺ (x = 0.25, 0.50, 0.75 and 1) is listed in Table 3. As seen from Table 3, the CIE coordinated tuned from near white towards ideal white emission (0.33, 0.33) with the enhancement in Gd³⁺ content and the co-ordinate (0.34, 0.36) was the nearest one. The CIE diagram of Y_0.25Gd_0.75VO₄:0.2% Dy³⁺ annealed at different temperatures is shown in S6 (ESI data†) and the obtained colour co-ordinates and CCT values are shown in S7 (ESI data†). There is no as such variation in colour co-ordinates with annealing temperature was observed. However with addition of Gd in the host there is a shift from blue (for YVO₄) to white can be noticed. Hence colour tuning in the prepared material is possible with changes in the host environment. The colour quality of white light in terms of colour correlated temperature (CCT) is given by the McCamy empirical formula

CCT = −437n³ + 3601n² − 6861n + 5514.31

where n = (x − x_e)/(y − y_e) and the chromaticity epicenter is at x_e = 0.3320 and y_e = 0.1858.⁷ The CCT values were found to vary from 5216–7636 K. Generally the CCT value greater than 5000 K is for cold white light emission which is used in commercial lighting. Present phosphors have the CIE and CCT values near the range of white light. So it can have possible application in cold white light emission in solid state lighting.

Table 3 CIE co-ordinates and CCT values

Y1−xGdxVO4:0.2% Dy^3+ x =	CIE co-ordinates		CCT (in K)
	X	Y
0	0.29	0.34	7636
0.25	0.30	0.35	6975
0.50	0.34	0.36	5216
0.75	0.31	0.35	6493
1.00	0.31	0.32	6737

---

4. Conclusions

A series of Y_1−xGd_xVO₄ doped with Dy³⁺ were prepared through simple co-precipitation technique and luminescence properties were investigated in detail. The structural studies were carried out for the confirmation of phase formation. FESEM and HRTEM images exhibited nano particle formation. SAED pattern confirmed the polycrystalline behaviour of the prepared sample. Optical band gap estimated from diffuse reflectance spectra by using Kubelka–Munk plot. Dy³⁺ doped Y_1−xGd_xVO₄ showed two prominent emission bands centred at 485 nm and 575 nm and the maximum intensity was obtained for Y_0.25Gd_0.75VO₄. Crystal environment played a vital role in optimization of emission of Dy³⁺. Effect of heat treatment on structural, morphological and optical properties were investigated which showed enhancement in crystalline nature and increment in particle size along with enhancement in PL excitation and emission bands. Photometric study revealed the near white emission property by calculating the CIE and CCT values. Colour tunability in the prepared samples is possible by varying the Gd³⁺ contents. For Y_0.50Gd_0.50VO₄:0.2% Dy³⁺ sample the evaluated CIE co-ordinate was (0.32, 0.34) with CCT value 5500 K. The obtained parameters assured its applicability in various display devices.

Acknowledgements

Puja Kumari gratefully acknowledges Indian School of Mines, Dhanbad for providing research fellowship funded by Government of India. The authors are also grateful to Dr S. K. Sharma department of Applied Physics for PL measurement and Dr S. Das, National Taiwan University Taipei for his continuous support throughout the work. The authors highly acknowledge Prof. S. B. Rai for the life time measurements.

References

1. B. Liu, C. Shi and Z. Qi, Appl. Phys. Lett., 2005, 86, 191111.
2. Z. Xia, J. Zhuang and L. Liao, Inorg. Chem., 2012, 51, 7202.
3. Q. Yin, X. Jin, G. Yang, C. Jiang, Z. Song and G. Sun, RSC Adv., 2014, 4, 53561.
4. N. T. Tran, J. P. You and F. G. Shi, J. Lightwave Technol., 2009, 27, 15.
5. S. Som, et al., Dalton Trans., 2014, 43, 9860.
6. T. V. Gavrilovic, D. J. Jovanovic, V. Lojpur and M. D. Dramićanin, Sci. Rep., 2014, 4, 4209.
7. P. Kumari and J. Manam, Spectrochim. Acta, Part A, 2016, 152, 109.
8. S. Das, C. Y. Yang and C. H. Lu, J. Am. Ceram. Soc., 2013, 96, 1602.
9. K. Park, S. W. Nam and M. H. Heo, Ceram. Int., 2010, 36, 1541.
10. T. Kim, et al., RSC Adv., 2014, 4, 45687.
11. H. Hu, et al., Biomaterials, 2014, 35, 8649.
12. L. Chen, K. J. Chen, S. F. Hu and R. S. Liu, J. Mater. Chem., 2011, 21, 3677.
13. E. Pavitra, G. S. R. Raju, W. Park and J. S. Yu, New J. Chem., 2014, 38, 163.
14. X. Zhang, F. Meng, W. Li, S. Kim, Y. M. Yu and H. J. Seo, J. Alloys Compd., 2013, 578, 72.
15. P. Kumari, P. K. Baitha and J. Manam, Indian J. Phys., 2015, 89, 1297.
16. K. Venkateswarlu, A. Chandra Bose and N. Rameshbabu, Phys. B, 2010, 405, 4256.
17. S. Das, C. C. Chang, C. Y. Yang, S. Som and C. H. Lu, Mater. Charact., 2015, 106, 25.
18. N. S. Singh, R. S. Ningthoujam, G. Phaomei, S. D. Singh, A. Vinud and R. K. Vatsa, Dalton Trans., 2012, 41, 4404.
19. W. Xu, et al., J. Phys. Chem. C, 2010, 114, 14018.
20. X.-H. Wang, P.-L. Chen and I.-W. Chen, Two-step sintering of ceramics with constant grain-size, I. Y₂O₃, J. Am. Ceram. Soc., 2006, 89, 431–437.
21. D. J. Jovanović, Opt. Mater., 2013, 35, 1797.
22. P. K. Baitha, P. P. Pal and J. Manam, Nucl. Instrum. Methods Phys. Res., Sect. A, 2014, 745, 91.
23. J. Tauc and A. Menth, J. Non-Cryst. Solids, 1972, 569, 8.
24. S. Khanahmadzadeh, M. Enhessari, Z. Solati, A. Mohebalizadeh and A. Alipouramjad, Mater. Sci. Semicond. Process., 2015, 31, 599.
25. S. Som, et al., Ultrason. Sonochem., 2016, 28, 79.
26. Q. Wang, Z. Zhang, Y. Zheng, W. Cai and Y. Yu, CrystEngComm, 2012, 14, 4786.
27. K. Venkata Rao, S. Babu, G. Venkataiah and Y. C. Ratnakaram, J. Mol. Struct., 2015, 1094, 274.
28. J. W. Stouwdam, M. Raudsepp and F. C. J. M. Veggel, Langmuir, 2005, 21, 7003.
29. B. K. Grandhe, V. R. Bandi, K. Jang, S. Ramaprabhu, S. S. Yi and J. H. Jeong, Electron. Mater. Lett., 2011, 7, 161.
30. C. Z. Peng, W. Y. Hua and Z. J. Chi, Chin. Phys. B, 2010, 19, 057803.
31. M. Gu, Q. Liu, S. Mao, D. Mao and C. Chang, Cryst. Growth Des., 2008, 8, 1423.
32. A. P. A. Marques, M. T. S. Tanaka, E. Longo, E. R. Leite and I. L. V. Rosa, J. Fluoresc., 2011, 21, 893.
33. C. S. Huang, Z. Cheng, P. Ma, X. Kang, Y. Daia and J. Lin, Dalton Trans., 2013, 42, 6523.
34. G. Wang, et al., J. Phys. Chem. C, 2008, 112, 17042.
35. M. N. Luwang, R. S. Ningthoujam, S. K. Srivastava and R. K. Vatsa, J. Mater. Chem., 2011, 21, 5326.
36. Q. Su, Z. Pei, L. Chi, H. Zhang, Z. Zhang and F. Zou, J. Alloys Compd., 1993, 192, 25.
37. F. Liu, Q. Liu, Y. Fangn, N. Zhang, B. Yang and G. Zhao, Ceram. Int., 2015, 41, 1917.
38. A. A. Reddy, et al., AIP Adv., 2013, 3, 022126.
39. C. Z. Peng, W. Yu-Hua and Z. Jia-Chi, Chin. Phys. B, 2010, 19, 057803.
40. P. F. S. Pereira, et al., J. Alloys Compd., 2012, 526, 11.
41. Y. Iso, S. Takeshita and T. Isobe, J. Phys. Chem. C, 2014, 118, 11006.
42. X. Liu, W. Hou, X. Yang and J. Liang, CrystEngComm, 2014, 16, 1268.
43. M. Chowdhury and S. K. Sharma, RSC Adv., 2015, 5, 51102.
44. S. Som, S. Dutta, V. Kumar, V. Kumar, H. C. Swart and S. K. Sharma, J. Lumin., 2014, 146, 162.

---

Footnote

† Electronic supplementary information (ESI) available: S1 to S7. See DOI: 10.1039/c5ra18982a

---