Effect of Annealing on Crystal-phase Transition and Photoluminescence of ZnS Nanoparticles
张小波,宋宏伟,于立新,孙宝娟,吕少哲,任新光,谢宜华,王晓君,曹健林
DOI: https://doi.org/10.3321/j.issn:1000-7032.2004.01.014
2004-01-01
Chinese Journal of Luminescence
Abstract:Since Bhargava firstly reported in 1994 that the luminescence quantum efficiency was up to 18% in (ZnS∶)Mn nanocrystalline, numerous studies on ZnS nanocrystalline were emerged. However, most of these studies focused on the luminescence of doped ZnS nanoparticles, but the study about the effect of annealing on crystal-phase transition and photoluminescence in ZnS nanoparticles was relatively rare. In this paper, some efforts were made. The cubic undoped and Eu-doped ZnS nanoparticles were prepared by co-precipitation, and the samples were achieved by the annealing of as-made nanoparticles at different temperatures. For the comparison, the phosphors were also made by the annealing of ZnS raw powders (bulk material for commercial use) under the same condition. During the annealing of nanoparticles, the particles grown up dramatically and the average diameter increased from 11 nm to 3～5 μm. The structural phase transition (from cubic to wurtzite) was studied by X-ray diffraction patterns and the experiment of difference thermal analysis (DTA). The result indicated that the phase transition temperature, in the samples annealed with ZnS nanoparticles, was 618 ℃, which decreased greatly due to surface effect than the value (800 ℃) reported by the reference. The luminescent properties of the samples were studied and compared to the phosphors annealed with ZnS raw powders. Two broad emission bands were observed in the samples annealed above 900 ℃. One was centered at 460 nm and the other at 520 nm. While in the phosphors annealed with ZnS raw powders, only 460 nm-centered emission was detected. The 520 nm-centered emission was irrelevant with the crystal phase transition and independent on the 460 nm-centered emission. So it maybe originated from different recombination centers. But in the measurement of time-resolved spectra and lifetime, they differed little from each other. However, it can be observed that the peak of the emission band of time-resolved spectra red-shifted as the delay time varied from 5～25 μs, which was the characteristics of recombination of donor-acceptor pairs. The normalized emission intensity decayed following with a bi-exponential fitting function, ηe~(-t/τ_1)+(1-η)e~(-t/τ_2). For 520 nm-centered emission, τ_1, τ_2 and η were determined to be 64, 250 μs and 0.47 respectively. For 460 nm-centered emission, τ_1, τ_2 and η were determined to be 68, 251 μs and 0.57. The decay time constants had little change with detected wavelength, but the ratio η changed with wavelength. The 460 nm-centered emission was generally ascribed to self-actived luminescence, and the 520 nm-centered emission was interpreted as the formation of new defaults induced by the preparation of nanoparticles or of some new dopants brought by the annealing. Doping of europium did not induce new luminescent centers, but caused the luminescent efficiency of defects to increase greatly.