Research Article www.acsami.org
Colossal Dielectric Behavior of Ga+Nb Co-Doped Rutile TiO2 Wen Dong,† Wanbiao Hu,*,† Adam Berlie,†,§ Kenny Lau,† Hua Chen,‡ Ray L. Withers,† and Yun Liu*,† †
Research School of Chemistry, ‡Center for Advanced Microscopy, The Australian National University, Canberra, Australian Capital Territory 2601, Australia § The Bragg Institute, Australian Nuclear Science and Technology Organisation, New Illawarra Road, Lucas Heights, Sydney, New South Wales 2234, Australia S Supporting Information *
ABSTRACT: Stimulated by the excellent colossal permittivity (CP) behavior achieved in In+Nb co-doped rutile TiO2, in this work we investigate the CP behavior of Ga and Nb co-doped rutile TiO2, i.e., (Ga0.5Nb0.5)xTi1−xO2, where Ga3+ is from the same group as In3+ but with a much smaller ionic radius. Colossal permittivity of up to 104−105 with an acceptably low dielectric loss (tan δ = 0.05−0.1) over broad frequency/temperature ranges is obtained at x = 0.5% after systematic synthesis optimizations. Systematic structural, defect, and dielectric characterizations suggest that multiple polarization mechanisms exist in this system: defect dipoles at low temperature (∼10−40 K), polaronlike electron hopping/transport at higher temperatures, and a surface barrier layer capacitor effect. Together these mechanisms contribute to the overall dielectric properties, especially apparent observed CP. We believe that this work provides comprehensive guidance for the design of new CP materials. KEYWORDS: colossal permittivity, dielectric properties, rutile TiO2, defect dipole, ceramic
1. INTRODUCTION Colossal permittivity (CP) (>103) materials are drawing increased attention because of their promising potential for applications in the areas of device miniaturization and energy storage.1 High-performance CP materials, however, are rare. Although a very high permittivity of up to 104−105 can be achieved in BaTiO3-based perovskite materials,2,3 they typically suffer from a strong temperature dependence of this permittivity, especially in the vicinity of their ferroelectric-toparaelectric phase transition. Other known CP materials, e.g., CaCu3Ti4O12 (CCTO)- and NiO-based systems, have relatively stable CP properties over a broad temperature range but poor dielectric losses, typically higher than 0.1.4−7 As a result, the main issue for the development of high-performance CP materials is the simultaneous lowering of the dielectric loss and improvement of the temperature/frequency stability of the dielectric properties. To this end, a new electron-pinned defect dipole mechanism, by means of which electrons can be localized within appropriately constructed defect clusters, was proposed,8 and In+Nb co-doped rutile TiO2 was cited as an example of this mechanism.8,9 This material exhibits both a large temperature-/frequency-independent CP (>104) as well as a low dielectric loss (100 kHz. Furthermore, over the same frequency range, a broad peak appears in the tan δ spectra. However, at a co-doping level of 0.5%, a relatively stable CP value over 104 accompanied by a relatively low tan δ (
