2018, Vol.47, No.10

1282-1284

The effects of the host crystal structure of an SrTa4O11:Er3+/Yb3+ up-conversion (UC) phosphor on its luminescence properties were investigated using the tetragonal tungsten bronze (TTB) and hexagonal polymorphs of the SrTa4O11 host. Hexagonal SrTa4O11:Er3+/Yb3+ showed bright luminescence, whereas the UC emission intensity of TTB SrTa4O11:Er3+/Yb3+ was very weak, corresponding to ≈1% that of the hexagonal phase. This result was attributed to the local crystal structure around the rare-earth ions in the host lattice.

Up-conversion (UC) phosphors convert two or three low-energy photons into one higher-energy photon by multistep excitation.1 Owing to this unique ability, these materials have been extensively investigated in recent years in biological imaging and therapies, security inks, 3D displays, and solar cells.26 In these applications, the UC luminescence intensity is a vital characteristic. Regrettably, UC phosphors have inherently low-luminescence efficiencies.

UC luminescent materials are typically formed from a host crystal and two different types of trivalent rare-earth ions (RE3+), i.e., a sensitizer and an emission center. In this system, the sensitizer absorbs near-infrared light. Yb3+ is typically chosen as the sensitizer because of its large absorption cross-section at 980 nm. The absorbed energy is then transferred to an emission center, such as Er3+, Ho3+, or Tm3+, which emits the up-converted visible light. UC luminescence originates from 4f–4f dipole transitions of RE3+ ions.7

In order to obtain brighter UC phosphors with higher emission efficiencies, knowledge of how the host structure affects luminescence properties is required. It is well known that UC luminescence properties are profoundly affected by the phonon energy of the host crystal. However, the mechanism by which the host crystal structure influences UC luminescence properties is not fully understood.

In this study, we investigated the influence of host crystal structure on the UC luminescence properties of SrTa4O11:Er3+/Yb3+ UC phosphors. Two crystal polymorphs of the SrTa4O11 host are known, i.e., tetragonal tungsten bronze (TTB) and hexagonal.8,9

TTB and hexagonal SrTa4O11:Er3+/Yb3+ were synthesized via the gelable citrate complex method10,11 and hydrothermal method, respectively.

A stock solution containing Sr2+ ions was prepared by dissolving Sr(NO3)2 (Kanto Chemical Co., Inc., 98.0%) in distilled water. Stock solutions containing Er3+ or Yb3+ ions were prepared from RE2O3 (where RE = Er or Yb, Japan Yttrium Co., Ltd., 99.9%). RE2O3 was added to nitric acid under magnetic stirring, and the excess nitric acid was removed by slow heating. Then, distilled water was added to obtain aqueous RE(NO3)3. Separately, TaCl5 (Furuuchi Chemical Corporation, 99.9%) was dissolved in a mixture of H2O2 and ammonia solution in an ice bath to form a peroxo-tantalum acid complex. Citric acid (CA) was added to this solution to stabilize the Ta complex, and the excess H2O2 and ammonia were removed by heating at 60 °C. Distilled water was then added to obtain a stock solution of the Ta–CA complex.

For the gelable citrate complex method, starting solutions of the constituent elements were mixed in a test tube. The molar ratio of (Sr:Er:Yb):Ta was (0.94:0.02:0.04):4. An aqueous CA solution, which acts as a complexing agent, was added at a metal/CA molar ratio of 1:5. The resultant transparent metal complex solutions were heated at 120 °C to form gels, calcined at 450 °C to obtain the precursor, and then heat treated at 1,200 °C for 5 h.

For the hydrothermal synthesis, the metal ratio was the same as above, and the total amount of metal was 2 mmol. Each starting solution and 10 mL of H2O2 were placed in a Teflon autoclave, and distilled water was then added to provide a total volume of 20 mL. After hydrothermal treatment at 230 °C for 20 h, the resulting product was separated by centrifugation and collected after repeated washing with distilled water. The powder sample was further calcined at 1,100 °C for 0.5 h to obtain the final crystalline oxide phosphor.

Powder X-ray diffraction analysis was performed using an X-ray diffractometer (D8 ADVANCE, Bruker Japan K.K., Tokyo, Japan) equipped with an X-ray tube (Cu Kα radiation, λ = 0.1540596 nm). The UC luminescence properties of the samples were examined using a variable-power 980 nm diode laser (L980P300J, THORLABS, NJ, USA) as the excitation source and a multichannel photodetector (MCPD-7700:311C, Otsuka Electronics Co., Ltd., Osaka, Japan). The down-conversion (DC) luminescence spectra were obtained on a spectrofluorometer (FP-8600, JASCO, Tokyo, Japan) with a 150 W xenon lamp as the excitation source. The absorption properties were investigated by spectrophotometry (V-670, JASCO Corporation, Tokyo, Japan).

The XRD patterns of the final products are shown in Figure 1. TTB SrTa4O11 was obtained using the gelable citrate complex method, whereas hexagonal SrTa4O11 was obtained by the hydrothermal method. Detailed crystallographic data for TTB SrTa4O11 are not available in the Inorganic Crystal Structure Database (ICSD). Therefore, the present report presents the diffraction pattern of tetragonal Ba5.5Ta21.8O60 (ICSD No. 24910) as a standard TTB pattern. No impurity phases from Er3+ or Yb3+ are observed in the samples, indicating that Er3+ and Yb3+ were successfully introduced into the host lattice. Hexagonal SrTa4O11 is a low-temperature phase. However, hexagonal samples were not obtained using the gelable citrate complex method even at low temperatures, as shown in Figure 2.

The UC emission spectra of TTB and hexagonal SrTa4O11 co-doped with 2% Er3+ and 4% Yb3+ were recorded under excitation at 980 nm using a laser diode. As shown in Figure 3, hexagonal SrTa4O11:Er3+/Yb3+ exhibits much stronger UC luminescence than TTB SrTa4O11:Er3+/Yb3+. The UC emission of the TTB phase is very weak, presenting a luminance that is ≈1% that of the hexagonal polymorph.

Figure 4 shows an energy level diagram of the possible transitions in an Er3+/Yb3+ co-doped system. For such systems, efficient UC luminescence is associated with an energy transfer (ET) UC mechanism.1,12,13 When the samples are excited at 980 nm, Yb3+ undergoes a 2F7/22F5/2 transition. The first ET from the Yb3+ ion in the 2F5/2 state to an Er3+ ion populates the 4I11/2 level of Er3+. In the second ET process, further excitation to the 4F7/2 level occurs. Subsequently, through non-radiative relaxation, these electrons decay to the 2H11/2, 4S3/2, and 4F9/2 levels, leading to green and red emissions. Alternatively, the 4I11/2 level of Er3+ can relax to the 4I13/2 state, and the 4F9/2 level can then be populated because of excitation from the 4I13/2 level.

Figure 5 shows the crystal structures of TTB Ba5.5Ta21.8O60 (as a reference for TTB SrTa4O11) and hexagonal SrTa4O11 and the coordination environments of the Sr2+ sites, where the RE3+ ions are located. The coordination environment for RE3+ is quite different between the TTB and hexagonal structures. The 4f–4f dipole transitions in RE3+ ions are sensitive to the local environment.14,15 The RE3+ ions in the TTB lattice occupy a site with a center of inversion, and, according to selection rules, the electric dipole transition is thus forbidden. In contrast, hexagonal SrTa4O11 crystals have low symmetry around the RE3+ ions in the lattice matrix. These low-symmetry conditions allow the occurrence of the 4f–4f transitions in the RE3+ ions, thus strengthening the emission intensity. Thus, UC emission intensity is strongly dictated by local crystal structure. The local crystal structure in hexagonal SrTa4O11 exhibits the same symmetry as that of YTa7O19, which shows an intense UC emission, as previously reported by our group.16

To further understand the effect of local crystal structure on UC luminescence properties, each process leading to UC emission was investigated. The diffuse reflectance spectra and the DC luminescence spectra under excitement at 490 nm are shown in Figures 6 and 7, respectively. The absorption by RE3+ in the excitation wavelength region (∼980 nm) for the TTB structure is 50% that of the hexagonal structure, but this difference is much smaller than the difference in UC luminescence intensity. Similarly, the DC emission intensity of the TTB structure is 20% lower than that of the hexagonal structure, but this difference is again much smaller than the difference in UC emissions for the two polymorphs. This indicates that the influence of local crystal structure in the absorption and emission processes is not sufficient to fully determine the UC luminescence characteristics. Furthermore, at 490 nm excitation, the 4F7/2 state of Er3+ is directly excited without going through the Yb3+ → Er3+ ET process. Thus, local crystal structure has a profound effect on the ET process. According to these results, the ET between Yb3+ occurs many times before Er3+ is populated at the emission level in the UC process.

In conclusion, we synthesized TTB and hexagonal SrTa4O11:Er3+/Yb3+ UC phosphors and investigated their UC luminescence properties. Hexagonal SrTa4O11:Er3+/Yb3+ shows dramatically stronger UC emission than that of TTB SrTa4O11:Er3+/Yb3+. This phenomenon is attributed to the low symmetry around the RE3+ ions in the hexagonal lattice, as the local crystal structure can have a significant influence on ET processes.

The authors acknowledge the support from the Research Program for Next-Generation Young Scientists of “Network Joint Research Center for Materials and Devices: Dynamic Alliance for Open Innovation Bridging Human, Environment and Materials” and JSPS KAKENHI Grant Number 16H02391 for financial support.