1. Introduction

Energy transfer between a donor and an acceptor has been demonstrated to enhance luminescence intensity via an antenna effect in many fields [1–16]. Energy transfer is important not only because it enhances emission intensity but also because it allows emission color tuning. Various kinds of inorganic phosphors employing energy transfer between RE cations have been examined, and while this is scientifically interesting, the net efficiency of emission is not sufficient if the donor ion possesses a forbidden excitation transition. Therefore, an emission center with high probability of transition to excited states is needed as a donor for practical applications. Considering an emission center exhibiting high transition probability, i.e., a parity-allowed excitation process, we can select two candidates for the donor: RE cations exhibiting 4f^x–4f^(x-1)5d transitions and ns²-type centers.

In the present study, we have focused on the ns²-type emission center as an activator. This emission is strongly affected by the coordination field, because the ns²-type centers (n ≧ 4) possess an electron in the outermost shell in both the ground state (ns²) and the excited state (ns¹np¹) [1, 17–21]. We have reported on the highest quantum efficiency (QE) in amorphous SnO–ZnO–P₂O₅ (SZP) low-melting glass [22–27]. It is notable that transparent glass without RE cations shows high UV-excited photoluminescence (PL) that is comparable to that of crystalline phosphor such as MgWO₄. Glass possessing strong light emission will be considered as a novel emitting material capable of good formability that is quite important in the industrial manufacturing process. The emission is caused by Sn²⁺, which is the most conventional and harmless ns²-type center. We have also reported that white light emission can be obtained by co-doping of Mn²⁺ in SZP glasses owing to the energy transfer from Sn²⁺ to Mn²⁺ centers [23]. White light emission was attained by the addition of Mn²⁺ cations, similar to the case of Mn²⁺-co-doped calcium halophosphate crystals containing Sb³⁺ centers [2].

Energy transfer from Sn²⁺ to Mn²⁺ in solid-state matter has been mainly reported for halide [10–12] and oxide crystal systems [4]. In previous reports, it was suggested that the emission centers of Sn²⁺ and Mn²⁺ were not randomly distributed but rather formed small Sn–Mn clusters inside alkali halides, such as NaCl, KBr, and NaBr, even when a very small amount was added. In addition, the energy transfer due to exchange interactions enables high energy-transfer rates. However, because the number of studies on the energy-transfer mechanism of Sn²⁺–Mn²⁺ in oxide glasses is limited [23], the details have not been clarified yet. We expect that the main reason for the lack of such a report is the valence control of the Sn²⁺ cation in glass materials. It was reported that the Sn²⁺ center in the SZP glass is easily oxidized into Sn⁴⁺ [26, 28]. Therefore, its preparation in an inert atmosphere is required in order to reveal the precise mechanism. Recently, our group compared the emission property of SZP glasses prepared in air and Ar atmospheres and confirmed that most of the Sn²⁺ centers were oxidized during melting in air. On the other hand, only Sn²⁺ centers existed in the SZP glass prepared in an Ar atmosphere [26], thereby confirming that the discussion on the energy-transfer mechanism is dealing with real Sn²⁺ centers in oxide glasses. Therefore, this work reports on the energy-transfer mechanism from Sn²⁺ to Mn²⁺ centers in ZnO–P₂O₅ glasses containing different amounts of Sn²⁺ centers (donors). For the present study, the chemical compositions of the samples were selected to be xMnO–ySnO–(60-y)ZnO–40P₂O₅ (in molar ratio). Herein, this glass system is denoted as xMn-ySZP.

2. Experimental

The xMn-ySZP glasses used in the present study were prepared according to the conventional melt-quenching method reported previously [26, 29]. The glass batch was set in a Ar-purged electric furnace at room temperature (r.t.). It took 2 h to heat up from r.t. to 1373 K; the temperature was then fixed at 1373 K for 30 min. After melting, the glass melt was quenched on a stainless steel plate at 473 K and then annealed for 1 h at the glass transition temperature (T_g) as measured by differential thermal analysis. Optical absorption spectra were measured using a U-3500 spectrophotometer (Hitachi). The PL and PL excitation (PLE) spectra were measured at r.t. by using an F-7000 fluorescence spectrophotometer (Hitachi). The emission decay at r.t. was measured using a Quantaurus-Tau instrument (Hamamatsu Photonics) with 280-nm LED. The absolute QE of the glass was measured using a Quantaurus-QY (Hamamatsu Photonics).

3. Results and discussion

Figure 1 shows the optical absorption spectra of the xMn-5.0SZP glasses. The observed absorption edges are due to the s–p transition of Sn²⁺ centers, and are slightly red-shifted because of the added MnO. Because such a shift of the absorption bands was not observed in xMn-1.0SZP and xMn-2.5SZP glasses, it is expected that the local coordination states of the Sn²⁺ centers in the xMn-5.0SZP glasses, whose Sn²⁺ concentration is far from the conventional concentration of crystalline phosphor, are affected by the Mn co-doping.

 

Fig. 1 Optical absorption spectra of xMn-5.0SZP glasses. The inset shows magnified spectra around absorption edges.

Download Full Size | PDF

Figure 2 shows the PL spectra of the xMn-5.0SZP glasses excited by a photon energy of 4.5 eV. These emission spectra consist of two broad bands of the donor Sn²⁺ (2.86 eV) and acceptor Mn²⁺ centers (2.05 eV) [23]. The emission intensity of Sn²⁺ decreases with increasing MnO amount, while that of Mn²⁺ increases in the concentration range less than 3.0 mol%. Such spectral changes due to energy transfer in glass were observed previously too [23]. The decrease in emission intensity in the case of Mn²⁺ addition above 5.0 mol% suggests concentration quenching of Mn²⁺. It is notable that Sn²⁺ emission is hardly observed in the 10Mn-5.0SZP glass, indicating that most of the excitation energy of the Sn²⁺ centers is transferred to Mn²⁺ in the present system. In order to examine the energy-transfer rate, we measured the emission decay curves of MnO-doped SZP glasses.

 

Fig. 2 PL spectra of xMn-5.0SZP glasses. The excitation energy was 4.5 eV.

Download Full Size | PDF

Figure 3(a) shows the PL emission decay curves of Sn²⁺ centers of, exemplarily, the xMn-5.0SZP glasses. A decrease in the decay constants of the Sn²⁺ centers with increasing Mn amount is observed. Because the energy-transfer process from the ns²-cation to Mn²⁺ in calcium halophosphate is thought to be governed by the exchange interaction [1], we take this interaction into account for the experimental fitting. The emission decay curves of the donor for the dominating exchange interaction are represented by the following equation [1, 31–33], which uses the radiative lifetime of the donor (τ_D):

 

Fig. 3 (a) PL emission decay curves of Sn²⁺ centers in xMnO-5.0SZP glasses (excitation at 4.43 eV and emission at 3.1 eV). Dots indicate experimental data and lines are the corresponding fitting curves obtained using Eq. (1). (b) τ_1/e values of xMnO-1SZP and xMnO-5.0SZP glasses as a function of the MnO amount.

Download Full Size | PDF

The function g(z) in Eq. (1) can be expressed by

The coefficients related to derivatives of the gamma function of argument unity, i.e., h₁, h₂, and h₃ are 1.73164699, 5.93433597, and 5.44487446, respectively [31]. The fitting curves in Fig. 3(a) show good agreement with the experimental data. From these fittings, we can obtain the τ_1/e values for the Sn²⁺ centers in the Mn-doped SZP glasses, shown in Fig. 3(b), for y = 1.0, 2.5, and 5.0. The emission decay constants of the glasses containing higher Sn concentrations exhibit slightly shorter lifetimes, suggesting concentration quenching between the Sn²⁺ centers.

If the decay constant of a Sn²⁺ center with a Mn²⁺ cation is denoted as τ, the energy-transfer efficiency η_ET is written according to [33–35] as

Figure 4(a) shows η_ET values for the xMn-ySZP glasses (y = 1.0, 2.5, and 5.0) as a function of the MnO amount. The values of η_ET from Sn²⁺ to Mn²⁺ generally increase with increasing Mn²⁺ amount. It seems that the η_ET values are independent of the SnO concentration. The present η_ET values are comparable to that reported in a previous study on Sn²⁺–Mn²⁺ phosphate glass prepared in air [23], although the Sn²⁺/(Sn²⁺ + Sn⁴⁺) ratio may be different. That is because no clustering of Sn²⁺ and Mn²⁺ occurs independent of the Sn²⁺ amount, which is also suggested by the present η_ET values. Figure 4(b) shows η_ET as a function of the Mn²⁺/Sn²⁺ ratio. It suggests that the distance between the donor and acceptor becomes statistically shortened. Since η_ET values depend on the Sn²⁺/Mn²⁺ ratio, the present data do not support the formation of Sn²⁺–Mn²⁺ clustering structures in the glasses prepared by the melt-quenching method. In other words, Sn²⁺ and Mn²⁺ centers are almost homogeneously dispersed in these glasses, like in a liquid state.

 

Fig. 4 Energy-transfer efficiency of xMnO-SZP glasses (a) as a function of the Mn²⁺ amount and (b) as a function of the Mn²⁺/Sn²⁺ ratio.

Download Full Size | PDF

The QE values of these glasses are shown in Fig. 5.It is notable that these glasses show a high QE value of around 90% even at high MnO doping. It is expected that the QE values may be influenced by several factors such as the coordination state or aggregation state of Sn and emission efficiency of Mn. The efficiencies obtained in the present study are more than twice the previously reported QE values for Sn²⁺–RE³⁺ co-doped glasses [27]. Therefore, these values clearly show the potential of the present MnO-doped SZP glasses for phosphor application with high QE. Although the generation of Sn²⁺–Mn²⁺ cluster-like structures has not been proven from the η_ET data, the QE values strongly suggest that effective energy-transfer paths exist even in a random matrix.

 

Fig. 5 QE values for xMnO-1SZP and xMnO-5.0SZP glasses as a function of the MnO amount.

Download Full Size | PDF

We emphasize that the present Sn²⁺–Mn²⁺ co-doped glasses have high QE values that are comparable to that of crystalline phosphor, although the host matrix possesses a random network. The present glass system is attractive not only from the viewpoint of a RE-free inorganic amorphous material possessing effective emission properties but also from the aspect of representing a dispersion of different kinds of emission centers in a random matrix. Because monolithic materials can reduce optical loss due to scattering at the interface, transparent emission materials will be applicable in novel industrial applications in combination with high-power deep UV LEDs.

4. Summary

We have demonstrated energy transfer from Sn²⁺ to Mn²⁺ centers in ZnO–P₂O₅ glasses. The optical absorption edge of the glasses containing higher Sn concentrations is affected by the MnO amount. The emission decay curves of Sn²⁺ can be fitted using the exchange interaction, and the energy transfer efficiency suggests that Sn²⁺ and Mn²⁺ centers are almost homogeneously dispersed in these glasses, like in a liquid state. The internal quantum efficiency of Sn²⁺–Mn²⁺ co-doped glasses indicates that effective energy-transfer paths exist even in a random matrix. The high energy transfer rate is also a notable advantage of the present Sn²⁺-doped phosphate glass phosphor.