Optical Study of (Nb0.5In0.5)0.02 Ti0.98O2 Crystals

Abstract

This work was a study of pure TiO2Rutile crystals, as well as Rutile crystals 2% co-doped with Indium and Niobium (2-NITO). There is much interest surrounding co-doped TiO2re-cently, with several papers published on ’colossal permittivity’ in the lower frequency ranges (10^2-10^6Hz range). The aim of this work was to study the optical and Raman modes of pure and co-doped crystals to determine the effects co-doping has on these modes. Infrared reflectance along with Raman Spectroscopy were used for this purpose. In order to determine the dielectric function from the infrared data, the Factorized Model and Kramers-Kronig analysis were used. Since TiO2has a tetragonal unit cell, infrared measurements of both the a and c axes of both doped and undoped crystals were done. Thea-axis is known to have 3 optical modes, whereas the c-axis only has one. However an additional mode was seen in all spectra, believed to be caused by anharmonicity. In addition, the 136cm−1mode observed in polycrystalline conductivity spectra of 5 and 10-NITO lines up directly with the A2u mode and the 793cm−1 mode also appears in single crystal TiO2, meaning these are not new modes. However the 447cm−1 and 654cm−1 modes do not appear in our data, and are likely a result of higher percentage co-doping. The effect of co-doping was observed to be an overall decrease in the reflectance of TiO2. We also observed sizable increases inγtofor all modes in 2-NITO. In addition, the dielectric permittivity decreases below the first phonon mode; suggesting that the enhanced permittivity observed at lower frequencies is not caused by co-doped changes to phonon modes. All expected Raman-active modes were observed, however due to poor data resolution some of the peak positions appear to be slightly different than previously measured. Our Raman spectra showed new structures at around 300cm−1and 700cm−1 in the (100) surface spectra, it is possible these are combination lines.