1. Introduction

White light-emitting diodes (WLEDs) have attracted attention due to the advantages of compactness, high brightness, long operating lifetime, environmental friendliness as well as good physicochemical stability [1]. Presently, the most prevalent method of fabricating WLEDs is to combine commercial blue LED chips with yellow phosphors such as the well-known Ce³⁺-doped yttrium aluminum garnet (YAG:Ce³⁺) [2]. However, this method has some inherent drawbacks, such as poor color rendering index and high color temperature [3]. Therefore, an alternative strategy based on a combination of a near ultraviolet (NUV) LED chip (380-410 nm) with red, green, and blue phosphors has been proposed to achieve a warm white light [4]. Compared with the blue and green phosphors, the commonly used Y₂O₂S:Eu³⁺ red phosphor shows lower physicochemical stability [5].

Recently, Eu³⁺-doped tungstate and molybdate phosphors have attracted much attention due to the advantages of easy-synthesis, high physicochemical stability, and intense red light emission [6, 7]. In order to strengthen the absorption of Eu³⁺ in the NUV region, some tungstates and molybdates that can be high Eu³⁺-doped have been investigated, such as Na₅Y(MoO₄)₄ and Li₃Ba₂La₃(MoO₄)₈ [8, 9]. Especially, it has been found that the concentration quenching effect in some defect scheelite structures is weak, such as the optimal Eu³⁺ concentrations in the M₂Gd₄(MoO₄)₇ (M = Li and Na) phosphors can be up to 85at.% and 70at.%, respectively [10]. LiGd₃(MoO₄)₅ crystal also has a similar defect scheelite structure. Compared with the scheelite CaWO₄, the Ca²⁺ sites in the LiGd₃(MoO₄)₅ are not fully occupied by Li⁺ and Gd³⁺, and 20% of them are vacancies [11].

In this work, LiGd₃(MoO₄)₅:Eu³⁺ red phosphors with different Eu³⁺ concentrations were synthesized by the solid state reaction method. The effects of Eu³⁺ concentration on the photoluminescence (PL) properties of the phosphors were investigated in detail. Meanwhile, the thermal stability, chromatic property, and quantum yield of the phosphors are reported.

2. Experimental

2.1 Preparations

A series of red phosphors LiGd_3(1-x)Eu_3x(MoO₄)₅ (LGM:xEu³⁺, x = 0, 10, 20, …, 100at.%) were synthesized by the solid-state reaction method. The starting materials Li₂CO₃ (Aladdin, A.R.), Gd₂O₃ (Aladdin, 99.99%), Eu₂O₃ (Aladdin, 99.99%) and MoO₃ (Aladdin, A.R.) were weighted according to the stoichiometric ratio. The raw materials were mixed and ground in an agate mortar thoroughly. Then, the homogenous mixtures were transferred to an alumina crucible and had been sintered in air at 500 °C for 3 h in a muffle furnace. The pre-samples were re-ground and had been calcined at 850 °C for 5 h again. After cooling to room temperature naturally, the as-prepared phosphors were ground into powders for measurements.

2.2 Characterizations

XRD measurements were carried out on a MiniFlex 600 powder diffractometer with Cu Kα radiation (1.5405 Å) operating at 40 kV and 15 mA. The scanning speed for phase determination was 0.2 °/min. The morphology and elemental analysis of the samples were inspected using a field emission scanning electron microscope (FESEM, SU-8010) equipped with an energy-dispersive spectrometer (EDS). Diffuse refection spectra (DRS) of the powder samples were measured by a spectrophotometer (Lambda900, Perkin-Elmer) and using BaSO₄ as a standard reference. Photoluminescence spectra, decay curves, and temperature-dependent luminescence spectra were recorded by a spectrometer (Edinburgh Instruments, FLS980) equipped with both continuous (450 W) and pulsed xenon lamps as the excited sources. Quantum yield (QY) was measured by a spectrometer (Edinburgh Instruments, FLS920) using a barium sulfate coated integrating sphere attached to the spectrometer.

3. Results and discussion

3.1 Phase composition

The recorded XRD patterns of the LGM:xEu³⁺ (x = 0, 10, 20, …, 100 at.%) phosphors are similar to each other, therefore, only three of them are shown in Fig. 1 for the sake of brevity. The main intense diffraction peaks of the samples can be well indexed by the tetragonal scheelite structure with space group I4₁/a as shown in Fig. 1(a). Besides, it can be found from the magnified patterns in Fig. 1(b) that there are some weak extra peaks that cannot be indexed by the scheelite structure. Similar peaks, which were also found in the recorded XRD patterns of Na_0.2Eu_0.6MoO₄, have been attributed to the cation-vacancy ordering and the formation of incommensurately modulated structure caused by the influence of a mass of vacancies in the crystal [12, 13]. The concrete distribution of the cations and vacancies has been solved in Ref [12]. Furthermore, it can be seen from Fig. 1(c) that the location of the diffraction peak corresponding to (112) slightly shifts to lower 2θ angel with the increment of Eu³⁺ concentration. According to the Bragg equation λ = 2dsinθ, the shift of peak location can be attributed to the slightly larger radius of Eu³⁺ (r = 1.066 Å, CN = 8) than that of Gd³⁺ (r = 1.053 Å, CN = 8) [14].

 

Fig. 1 (a) XRD patterns of the LGM:xEu³⁺ (x = 0, 50, and 90 at.%) phosphors, and the corresponding magnified XRD curves in the ranges of 10-28° (b) and 28-29.5° (c).

Download Full Size | PDF

The morphology and elemental compositions of the LGM:90at.%Eu³⁺ phosphor sample are shown in Figs. 2(a) and 2(b), respectively. The morphology in Fig. 2(a) shows that the sizes of the irregular particles are mainly ranging from 2 to 5 μm. As presented in Fig. 2(b), a small area of sample was selected for the EDS measurement. The result confirms the presence of Mo, Eu, and Gd elements and no other impurity peaks. However, the light Li element is hard to be identified by the EDS measurement. The atomic ratios of Eu/(Gd + Eu) and (Gd + Eu)/Mo in the sample were calculated to be about 90% and 58%, respectively, which are almost consistent with the chemical formula of Li(Gd_0.1Eu_0.9)₃(MoO₄)₅. To further investigate the spatial distribution of elements, the EDS elemental mapping analysis corresponding to the particle imaged in Fig. 3(a) was carried out, as shown in Fig. 3(b), (c) and (d). The pictures show that Mo, Eu, and Gd uniformly distribute and the as-prepared phosphor is homogeneous phase.

 

Fig. 2 (a) SEM micrograph and (b) EDS data of the LGM:90at.%Eu³⁺ phosphor.

Download Full Size | PDF

 

Fig. 3 The mapping EDS images of Mo (b), Eu (c) and Gd (d) corresponding to the particle shown in (a).

Download Full Size | PDF

3.2 Reflection and photoluminescence properties of Eu³⁺ doped LiGd₃(MoO₄)₅

Figure 4(a) shows the room-temperature diffuse reflection spectra of the LGM:xEu³⁺ (x = 0, 50, and 90 at.%) phosphors in 200-800 nm. The host LGM has a broad absorption band from 250 to 400 nm corresponding to the charge-transfer state (CTS) band of Mo⁶⁺-O^2- [15]. When Eu³⁺ ions are doped into the LGM, the CTS of Eu³⁺-O^2- appears and the absorption edge of Mo⁶⁺-O^2- CTS band shifts to a longer wavelength, which may be caused by the increment of covalency of Mo-O bond due to the incorporation of Eu³⁺ [5, 16]. Due to the larger electronegativity of Eu (1.18 [17]) than that of Gd (0.9 [18]), the electronegativity difference between Eu and O is smaller than that of Gd and O and the covalency of Eu-O bond is stronger than that of Gd-O bond. The stronger covalency of Ln-O bond in Ln–O–Mo will weaken the bond ionicity and increase the bond covalency of Mo–O bond. So the incorporation of Eu³⁺ results in the increment of covalency of Mo-O bond and the location of CTS of Mo⁶⁺-O^2- shifts to lower energy. Meanwhile, several narrow absorption bands can be observed in the visible region, which are originated from the characteristic absorption of Eu³⁺ intra-4f transitions. Obviously, with the increment of Eu³⁺ concentration, the characteristic absorption of Eu³⁺ is prominently enhanced, and the Mo⁶⁺-O^2- CTS leads to a stronger and broader absorption band in NUV region.

 

Fig. 4 (a) Room-temperature diffuse reflection spectra of the LGM, LGM:50at.%Eu³⁺ and LGM:90at.%Eu³⁺ phosphors in 200-800 nm, (b) Room-temperature excitation spectra of the LGM:10at.%Eu³⁺ and LGM:90at.%Eu³⁺phosphors.

Download Full Size | PDF

Figure 4(b) shows the room-temperature excitation spectra of the LGM:10at.%Eu³⁺ and LGM:90at.%Eu³⁺ phosphors monitored at 615 nm corresponding to the ⁵D₀→⁷F₂ transition. The excitation spectrum can be divided into two regions. One part is a broad band from 250 to 350 nm, which is mainly attributed to Eu³⁺-O^2- CTS band due to the intensity increases with increasing Eu³⁺ concentration [19]. The charge transfer absorption of Mo⁶⁺-O^2- existed in diffuse reflection spectra in NUV region may not make contribution to the ⁵D₀→⁷F₂ emission [10]. The other part includes sharp lines in 350-550 nm corresponding to the intra-4f transitions of Eu³⁺. Among these lines, the strongest exciting peak is located at 395 nm and assigned to the ⁷F₀→⁵L₆ transition, which implies that the LGM:Eu³⁺ phosphor might be used as red phosphor excited by the NUV LED chips.

Figure 5(a) shows the room-temperature emission spectra of the LGM:xEu³⁺ phosphors under excitation at 395 nm. The profile of emission spectrum is not changed with the Eu³⁺ concentration in the phosphors. In the emission spectrum, the host emission corresponding to Mo⁶⁺-O^2- CTS is not observed, which may be due to that thermal quenching occurred at room temperature [19]. The strongest emission peak of the phosphor at 615 nm is attributed to the ⁵D₀→⁷F₂ transition of Eu³⁺. The emissions originated from the ⁵D₀→⁷F₁ (592 nm), ⁵D₀→⁷F₃ (654 nm) and ⁵D₀→⁷F₄ (701 nm) transitions are much weaker. The ⁵D₀→⁷F₂ is a hypersensitive transition and strongly influenced by the coordination environment of Eu³⁺, while the ⁵D₀→⁷F₁ transition is insensitive to that [20]. Therefore, the emission intensity ratio R between the ⁵D₀→⁷F₂ and ⁵D₀→⁷F₁ transitions can be used as an index to measure the site symmetry of Eu³⁺ ions, and a larger R value means that the Eu³⁺ site in a host has a lower local symmetry [21]. In the LGM:xEu³⁺ phosphors, the intensity of the ⁵D₀→⁷F₂ is much stronger than that of the ⁵D₀→⁷F₁, and the R values are almost kept as a constant of 13.8 for different Eu³⁺ concentrations, as shown in Fig. 5(b). The value is smaller than that of the Li₂Gd₄(MoO₄)₇:85at.%Eu³⁺ (16 [10]), but larger than those of the Li₃Ba₂La₃(MoO₄)₈:90at.%Eu³⁺ (8.82 [8]) and KBaEu(MoO₄)₃ (12.4 [22]) phosphors. It indicates that the LGM:xEu³⁺ phosphor has a low Eu³⁺ site symmetry and high red color purity [22].

 

Fig. 5 (a) Room-temperature emission spectra of the LGM:xEu³⁺ (x = 10, 50, 90 at.%) phosphors excited at 395 nm. and the enlarged region in the inset. (b) the dependence of integrated emission intensity of the ⁵D₀→⁷F₂ transition and the (⁵D₀→⁷F₂)/(⁵D₀→⁷F₁) emission ratio R on the Eu³⁺ concentration.

Download Full Size | PDF

Figure 5(b) shows the integrated emission intensity of the ⁵D₀→⁷F₂ transition as a function of Eu³⁺ concentration. The emission intensity is increased with the increment of Eu³⁺ concentration and reached the maximum at x = 90 at.% (~68 × 10²⁰ cm⁻³). The optimal Eu³⁺ concentration in the LGM phosphors is higher than those in the M₂Gd₄(MoO₄)₇ (~62 × 10²⁰ and ~51 × 10²⁰ cm⁻³ for M = Li and Na, respectively [10]), KBaEu(MoO₄)₃ (~37 × 10²⁰ cm⁻³ [22]), Li₃Ba₂La₃(WO₄)₈ (~40 × 10²⁰ cm⁻³ [8]) and LiGd(MoO₄)₂ (~19 × 10²⁰ cm⁻³ [23]) phosphors. The result shows that the concentration quenching effect in the LGM:Eu³⁺ phosphor is very weak, which may be caused by its defect scheelite structure. The weak concentration quenching has also been found in some phosphors with the incommensurately modulated structure, such as Na₂Gd₄(MoO₄)₇:Eu³⁺ [24, 25] and Na_0.2Eu_0.6MoO₄ [12]. In the LGM:Eu³⁺ phosphor, Gd³⁺ ions may tend to occupy the adjacent Ca²⁺ sites in the scheelite structure and then gather together to form Gd-aggregates with longer distance between each other, as in the Na_0.2Eu_0.6MoO₄ phosphor [12]. Therefore, the cluster structure of Gd³⁺ in the LGM, which is similar to the layer structure of La³⁺ in Li₃Ba₂La₃(WO₄)₈ [8] and the chain structure of Eu³⁺ in KBaEu(MoO₄)₃ [22], makes the interaction between doping Eu³⁺ mainly occur inside the aggregates, and then reduces the energy migration to the quenching centers such as crystal defects or trace ions. On account of the fact that the cross-relaxation between Eu³⁺ ions is very weak [26], the concentration quenching should be mainly attributed to the energy migration from Eu³⁺ luminescent centers to the quenching centers [27]. Consequently, the concentration quenching effect is weak in the LGM:Eu³⁺ phosphor.

The excitation and emission spectra of the LGM:90at.%Eu³⁺ and the commercial Y₂O₂S:6.3at.%Eu³⁺ red phosphors recorded in the same experimental conditions are compared in Figs. 6(a) and 6(b), respectively. The Y₂O₂S:6.3at.%Eu³⁺ phosphor was bought from Jiangmen Kanhoo Industry Co., Ltd. It can be seen from Fig. 6(a) that the Y₂O₂S:6.3at.%Eu³⁺ phosphor shows a stronger and broader excitation band from 250 to 380 nm corresponding to the CTS of Eu³⁺-O^2- and Eu³⁺-S^2- [28], but the LGM:90at.%Eu³⁺ phosphor has a more efficient absorption in NUV region. Under excitation at 395 nm, the integrated emission intensity of the ⁵D₀→⁷F₂ transition for the LGM:90at.%Eu³⁺ phosphor is approximate 2.5 times of that of the Y₂O₂S:6.3at.%Eu³⁺ phosphor. Furthermore, the emission intensity ratio R between the ⁵D₀→⁷F₂ and ⁵D₀→⁷F₁ transitions for the Y₂O₂S:6.3at.%Eu³⁺ was calculated to be 7.9, and the LGM:90at.%Eu³⁺ has larger R value and narrower emission band corresponding to the ⁵D₀→⁷F₂ transition than the Y₂O₂S:6.3at.%Eu³⁺ phosphor, which can make the LGM:90at.%Eu³⁺ phosphor have higher color purity. The above results indicate that compared with the commercial Y₂O₂S:6.3at.%Eu³⁺ red phosphor, the LGM:90at.%Eu³⁺ is an excellent red one with higher luminescence intensity and color purity excited by the 395 nm LEDs. The LEDs with emission wavelength around 400 nm has the highest external quantum efficiency and is safer for human eyes among the LED chips with emission wavelength from 250 to 400 nm [29, 30]. It must be pointed out that the full width at half maximum (FWHM) of the exciting band around 395 nm for the LGM:90at.%Eu³⁺ phosphor is only 1.5 nm, therefore, On account of utilization ratio of excitation source, the LGM:Eu³⁺ phosphor may be more suitable for being excited by laser diode (LD) at 395 nm with narrower bandwidth. Efficient white light has been achieved by combining a NUV LD with red, green, and blue phosphors [31].

 

Fig. 6 (a) Excitation and (b) emission spectra of the LGM:90at.%Eu³⁺ and commercial Y₂O₂S:6.3at.%Eu³⁺ red phosphors.

Download Full Size | PDF

Figure 7 shows the fluorescence decay curves of the LGM:xEu³⁺ phosphors excited at 395 nm and the emission related to the ⁵D₀→⁷F₂ transition at 615 nm was monitored. All curves show a single exponential decay and the fluorescence lifetime τ of the ⁵D₀ level can be obtained by

 

Fig. 7 Fluorescence decay curves of the LGM:xEu³⁺ (x = 10, 30, 50, 70, 80, 90, and 100 at.%) phosphors.

Download Full Size | PDF

3.3 Thermal quenching property of LGM:90at.%Eu³⁺

The temperature dependent emission spectra of the LGM:90at.%Eu³⁺ phosphor under excitation at 395 nm are shown in Fig. 8. The emission intensity of the phosphor obviously declines with the temperature increment from 300 K to 550 K. As the temperature reaches to 450 K, the emission intensity declines to 58% of the initial value at 300 K. For Eu³⁺-doped tungstate and molybdate phosphors [8, 22, 32], the thermal quenching of fluorescence emitted from the ⁵D₀ level is usually owed to nonradiative decay via the charge transfer state that has a crossover point with the ⁵D₀ and ⁷F_J levels, as shown in Fig. 9(a) [33].

 

Fig. 8 Temperature dependence of fluorescence for the LGM:90at.%Eu³⁺ phosphor under excitation at 395 nm. The inset shows the temperature dependence of emission intensity of the ⁵D₀→⁷F₂ transition under excitation at 395 nm and 465 nm, respectively.

Download Full Size | PDF

 

Fig. 9 (a) Pathway for the thermal quenching of the ⁵D₀ state through a CTS band, (b) The dependence of ln(I₀/I-1) on 1/kT for the LGM:90at.%Eu³⁺ phosphor.

Download Full Size | PDF

The activation energy ∆E is the energy separation from the bottom of the ⁵D₀ level to the ⁵D₀-charge transfer state crossover, and can be calculated by the relationship of the emission intensity and temperature [34],

3.4 CIE chromaticity coordinates and quantum efficiency of LGM:90at.%Eu³⁺

The color chromaticity is regarded as a critical parameter for evaluating the performance of WLED phosphors. The corresponding Commission International del’Eclairage (CIE) chromaticity coordinate of the LGM:90at.%Eu³⁺ phosphor at room temperature is about (0.6705, 0.3292), as shown in Fig. 10, which is close to the standard red chromaticity (0.67, 0.33) for the National Television Standard Committee (NTSC) system. With the increment of temperature, the CIE chromaticity coordinate for the LGM:90at.%Eu³⁺ phosphor only changes slightly as shown in the inset of Fig. 10. According to Ref [38], the color purity describes the saturation and degree of vividness of color, and the closer to spectral locus the chromaticity coordinate is, the higher the color purity is. From Fig. 10, it can be clear that the LGM:90at.%Eu³⁺ phosphor indeed shows higher color purity than the Y₂O₂S:6.3at.%Eu³⁺ phosphor.

 

Fig. 10 CIE chromaticity coordinate of the LGM:90at.%Eu³⁺ phosphor at 395 nm excitation.

Download Full Size | PDF

The QY of the LGM:90at.%Eu³⁺ phosphor under excitation at 395 nm is 41%, which is higher than those of the KBaEu(MoO₄)₃ (16% at λ_ex = 394 nm [22]) and Na₂Gd₄(MoO₄)₇:70at.%Eu³⁺ (19% at λ_ex = 396 nm, 22.6% at λ_ex = 465 nm [10]), but lower than those of the Y₂O₂S:6.3at.%Eu³⁺ (76% at λ_ex = 317 nm, 77% at λ_ex = 395 nm, measured in our lab), and Li₃Ba₂La₃(WO₄)₈:90at.%Eu³⁺ (99% at λ_ex = 395 nm [8]) phosphors. The QY of the LGM:90at.%Eu³⁺ phosphor at 465 nm excitation is 87%, which is much higher than that at 395 nm excitation. Because there is no CTS interference at 465 nm excitation, so the low QY of the LGM:90at.%Eu³⁺ phosphor is likely due to that the charge transfer absorption of Mo⁶⁺-O^2- around 395 nm only little make contribution to the ⁵D₀→⁷F₂ emission. By partial substitution of W⁶⁺ for Mo⁶⁺, the concentration of the MoO₄ group is diluted. the distance between MoO₄ group becomes longer, and the energy transition is blocked by MoO₄, which may result in more energy transferred from the host to Eu³⁺ ions [39].

4. Conclusion

A series of LGM:xEu³⁺ red phosphors have been synthesized by the solid state reaction method. The variations of fluorescence spectra and lifetimes with the Eu³⁺ concentration show that the concentration quenching effect is weak and the optimal Eu³⁺ concentration can be up to 90at.% in the phosphors. The emission intensity of the LGM:90at.%Eu³⁺ phosphor is 2.5 times of that of the Y₂O₂S:6.3%Eu³⁺ commercial red phosphor under excitation at 395 nm. The LGM:90at.%Eu³⁺ phosphor has a high color purity with CIE chromaticity coordinates of (0.6705, 0.3292), which is very close to the NTSC system standard for red chromaticity. Therefore, the LGM:90at.%Eu³⁺ phosphor has potential applications for the NUV excited WLEDs, especially, laser-driving white lighting.