1. Introduction

Over the past decade, white light-emitting diodes (W-LEDs) have attracted considerable interest for their potential applications in lighting and display [1–3]. The growing market share of W-LEDs for solid state lighting impelled the great progress of III-Nitride based LEDs [4–13] and phosphor materials [14–16] for W-LEDs. So far, many methods to realize white light emission were developed [3,17,18]. The most common one is combining a blue-emitting GaN-based semiconductor chip with yellow-emitting phosphor [17–19]. However, the difference between the individual degradation rates of the blue LED chip and the phosphors results in chromaticity aberration and poor white light performance [14]. To solve this problem, red, green, and blue multi-phased phosphors are employed to generate white light under ultraviolet (UV) LED chips excitation. However, blue emission efficiency is low in this three-converter system because of the strong reabsorption of blue light by the red or green phosphors [18]. A large number of researchers have recently reported single-phased phosphors based on UV LED chips, which can avoid the aforementioned problems [20,21].

In our previous work, we presented a new single-phased γ-irradiated Ag/Eu co-doped phosphate glass phosphor based on near ultraviolet (NUV) LED chips, which offers an alternative for efficient white light emission [22]. The formation mechanism of silver aggregates (Ag₂⁺ and Ag₃²⁺) and Eu²⁺ ions; the efficient energy transfers of silver aggregates→Eu²⁺, silver aggregates→Eu³⁺, and Eu²⁺→Eu³⁺; and the effect of energy transfer on emission properties of γ-irradiated Ag/Eu co-doped phosphate glass were discussed.

In this study, we focus on the effects of Ag/Eu co-doping ratio and excitation wavelengths on white light emission. As a result, tunable efficient white light emission is obtained by simply adjusting Eu/Ag co-doping concentration and excitation wavelengths. Furthermore, the most suitable glass composition and excitation wavelength for white light emission are found.

2. Experimental

Phosphate glasses with compositions (mol%) of 65NaPO₃-30Al(PO₃)₃-5SiO₂-xAgNO₃-yEu₂O₃ (x = 1.5, y = 0.2, 0.5, 0.8, 1.2, 1.5, and x = 1, 2, 3, 4, y = 0.8) were melted via the traditional melt quenching method using reagent grade NaPO₃, Al(PO₃)₃, SiO₂, AgNO₃, and Eu₂O₃ as starting materials [22]. The stoichiometric chemicals were thoroughly mixed and melted in a covered alumina crucible at 1250 °C for 2 h. The melt was then cast into a preheated iron mold and annealed in a muffle furnace at 450 °C for 2 h. Samples with sizes of 25 × 10 × 2 mm³ were cut and polished for optical measurements. The selected glass samples were irradiated by γ-rays from ⁶⁰Co. The dose was 3 KGy, and the dose rate was 20.8 Gy/min. These samples were then annealed in a muffle furnace at 200 °C for 40 min to stabilize the luminescence centers. Electron paramagnetic resonance (EPR) spectra were collected on an EPR spectrometer operating at the X-band (υ = 9.084 GHz). The optical absorption spectra were recorded on a JASCO V-570 spectrophotometer in the spectral range of 200 nm to 800 nm. The emission spectra were measured on a FLS 920 fluorescence spectrophotometer with a 450 W Xe lamp as the excitation source. The Emission spectra were corrected by the spectral response of the detector. All analyses were performed at room temperature.

3. Results and discussion

The EPR [Fig. 1(a) ] and absorption [Fig. 1(b)] spectra were used to confirm the appearance of Eu²⁺ and silver aggregates after γ-irradiation. In our previous work [22], we confirmed the absence of Eu²⁺ and silver aggregates in glasses before γ-irradiation. The characteristic “U” spectra (at g~2.0, g~2.8, and g~6.0) of Eu²⁺ in EPR spectra suggest the partial reduction of Eu³⁺ after γ-irradiation [22–24]. The EPR signals of Ag²⁺ (g_⊥~2.05 and g_∥~2.35), Ag₂⁺ (g~1.994), and Ag₃²⁺ (g~1.980) were clearly observed in Fig. 1(a) [22,25]. Likewise, the absorption bands of Ag²⁺ at 260 nm, Ag₂⁺ at 320 nm, Ag⁰ at 345 nm, and Ag₃²⁺ at 375 nm were observed in Fig. 1(b) [25–27]. These results indicate the appearance of silver aggregates after γ-irradiation. The intensity of EPR signals and absorption bands of Eu²⁺ ions become stronger, whereas those of silver aggregates (Ag₂⁺ and Ag₃²⁺) become weaker with increasing Eu-doping concentration. This result is attributed to the competition between Eu³⁺ and Ag⁺ ions in trapping electrons. In γ-irradiated Ag/Eu co-doped glass, Eu³⁺ can trap irradiation induced electrons to form Eu²⁺ and Ag⁺ can trap electrons to form silver aggregates (Ag₂⁺ and Ag₃²⁺) [22,24]. The irradiation induced electrons are invariable for the same glass host under irradiation with identical dose. Therefore, When the Eu-doping concentration increases, more Eu³⁺ ions will trap electrons and the electrons trapped by Ag⁺ will decrease. As a result, the amount of Eu²⁺ increases while that of silver aggregates decreases with the increase of Eu-doping concentration. Thus, the increase in Eu-doping concentration suppresses the formation of silver aggregates. The strong absorption of silver aggregates in the NUV range indicates that NUV excitation light can be efficiently absorbed.

 

Fig. 1 (a): EPR and (b): absorption spectra of γ-irradiated Ag/Eu co-doped phosphate glasses. These glasses were doped with 1.5 mol% Ag and co-doped with 0.2, 0.5, 0.8, 1.2 and 1.5 mol% Eu. Inset of (b): zoom at the spectra region of 450-600 nm.

Download Full Size | PDF

Figure 2(a) shows the emission spectra of γ-irradiated Ag/Eu co-doped phosphate glasses as a function of Eu-doping concentration. The optimal excitation wavelength at 375 nm was used to record the emission spectra [22]. The corresponding broad emission band from 400 nm to 740 nm almost covers the entire visible range and is comprised of three parts. Three sharp emission peaks located at 590, 612, and 700 nm can be readily assigned to the ⁵D₀→⁷F_J transitions (J = 1, 2, and 4, respectively) of Eu³⁺ [16,28,29]. The broad shoulder centered at around 620 nm is attributed to γ-irradiation induced silver aggregates [22,30]. The broad blue emission band with a maximum at about 450 nm is attributed to Eu²⁺ [22]. As previously mentioned, the amount of silver aggregates slightly decreases with increasing Eu-doping concentration. Thus, the energy transfers from silver aggregates to Eu²⁺ and Eu³⁺ ions become weaker. As a result, the change in Eu-doping concentration is the dominant factor facilitating the change in the emission spectra. For the blue emission band of Eu²⁺, the emission intensity increases with Eu-doping concentration and attains a maximum at 0.8 mol% Eu. For higher Eu-doping concentrations, the emission intensity is significantly decreased due to concentration quenching [31–33]. For the red emission of Eu³⁺, the emission intensity increases with Eu-doping concentration and is unaffected by the concentration quenching effect.

 

Fig. 2 Emission spectra of γ-irradiated phosphate glasses (a): doped with 1.5 mol% Ag and co-doped with 0.2, 0.5, 0.8, 1.2 and 1.5 mol% Eu under 375 nm excitation; (b): integrated emission intensity as a function of Eu-doping concentration; (c): doped with 0.8 mol% Eu and co-doped with 1, 2, 3, and 4 mol% Ag under 375 nm excitation; and (d): doped with 0.8 mol% Eu and co-doped with 4 mol% Ag under excitation at different wavelengths.

Download Full Size | PDF

Figure 2(b) shows integrated emission intensity as a function of Eu-doping concentration. The integrated emission intensity over the entire emission spectra significantly increases with Eu-doping concentration. When Eu-doping concentration is 0.8 mol%, the integrated emission intensity is at its maximum. The integrated emission intensity significantly decreases when Eu-doping concentration is further increased. This result indicates that the optimal doping concentration of Eu for white light emission is approximately 0.8 mol%.

Thus, we fixed the Eu-doping concentration at 0.8 mol% and changed the Ag-doping concentration to study the effect of the latter on emission property. Figure 2(c) shows the corresponding emission spectra of γ-irradiated Ag/Eu co-doped glass samples as a function of Ag-doping concentration. Both the blue light emission of Eu²⁺ and the red light emission of Eu³⁺ increase with the increase of Ag-doping concentration, confirming the efficient energy transfers from silver aggregates to Eu²⁺ and Eu³⁺. In our previous work [22], we discussed the energy transfer from silver aggregates to Eu²⁺ and Eu³⁺ in detail. Because irradiation induced silver aggregates increases with the increase of Ag-doping concentration, the increase of Ag-doping concentration will enhance the energy transfer probability from silver aggregates to Eu²⁺ and Eu³⁺ and the emission intensity of Eu²⁺ and Eu³⁺ increases. Meanwhile, the peak of blue emission was slightly blue-shifted from 480 nm to 470 nm. The red emission intensity of silver aggregates increases but not as strong as that of Eu²⁺ with the increase of Ag doping concentration due to the concentration quenching and energy transfer from silver aggregates to Eu²⁺ and Eu³⁺. Thus, the emission intensity ratio between Eu²⁺ and Silver aggregates increases leading to the blue shifting of the blue emission band from 480 to 470 nm. For higher Ag-doping concentrations, that is, > 4 mol%, obtaining an efficient emission is theoretically possible, but glass formation is no longer possible because of the precipitation of silver atoms resulting in the devitrification of glasses. Thus, the optimal Ag-doping concentration for white light emission is approximately 4 mol% for the present glass system.

The emission property of the optimized sample (4Ag/0.8Eu) under different excitation wavelengths was also studied, as shown in Fig. 2(d). The blue light emission of Eu²⁺ and the red light emission of Eu³⁺ and silver aggregates indicates that the corresponding emission color can be continuously tuned by simply changing the excitation wavelengths. All excitation wavelengths from 335 nm to 385 nm, which matches the NUV LEDs well, show high emission efficiency. The emission spectra of the optimized glass samples (4Ag/0.8Eu) under excitation at different wavelengths were converted to the CIE 1931 chromaticity diagram, as shown in Fig. 3 . Upon different excitation wavelengths, all emission chromaticity coordinates approached the standard white light emission (X = 0.333, Y = 0.333). A pure white emission with the chromaticity coordinates (X = 0.331, Y = 0.329) was generated under excitation at 360 nm.

 

Fig. 3 CIE chromaticity diagram for the optimized phosphate glass doped with 0.8 mol% Eu and co-doped with 4 mol% Ag under different excitation wavelengths.

Download Full Size | PDF

Therefore, because of the broad covering the entire visible range emission band, for the propose of W-LEDs application, the single-phased Ag/Eu co-doped phosphate glass phosphor based on UV-LEDs has higher color rendering index (CRI) than the method combining a blue LED with yellow phosphor (Ce³⁺:YAG). Furthermore, the color stability is also higher than that of the latter since the emission color doesn’t change with the input power. For white light emission base on UV-LEDs, the main disadvantage is the low light conversion efficiency. Thus, it is important to develop high efficiency UV-LEDs and phosphor materials for W-LEDs using UV-LEDs as excitation resources.

4. Conclusions

In summary, we present a new white light emission of γ-irradiated Ag/Eu co-doped phosphate glass phosphor. The optimized Ag/Eu co-doped phosphate glasses for white light emission was achieved at a concentration of 4 mol% Ag and 0.8 mol% Eu. The white light emission can be continuously tuned by simply varying the Ag and Eu co-doping ratio and excitation wavelengths. A relative pure white emission with the chromaticity coordinates (X = 0.331, Y = 0.329) was generated for the optimized sample (Ag/Eu = 4/0.8) under excitation at 360 nm. This work suggests that γ-irradiated Ag/Eu co-doped phosphate glass has potential application in W-LEDs.