Oxidation of metastable single-phase polycrystalline Ti0.5Al0.5N films: kinetics and mechanisms

• Published in: Journal of applied physics 1990-02, Vol.67 (3), p.1542-1553
• Main Authors: MCINTYRE, D, GREENE, J. E, HAKANSSON, G, SUNDGREN, J.-E, MUNZ, W.-D
• Format: Article
• Language: English
• Subjects: Condensed matter: structure, mechanical and thermal properties; Equations of state, phase equilibria, and phase transitions; Exact sciences and technology; Other topics in equations of state, phase equilibria, and phase transitions; Physics
• ISSN: 0021-8979; 1089-7550
• DOI: 10.1063/1.345664

Metastable single-phase, NaCl-structure, polycrystalline Ti0.5Al0.5N alloy films have been shown to exhibit much better high-temperature (750–900 °C) oxidation resistance than polycrystalline TiN films grown under similar conditions. The Ti0.5Al0.5N alloys, ≂3 μm thick, were deposited at temperatures between 400 and 500 °C on stainless-steel substrates by dc magnetron sputter deposition in mixed Ar+N2 discharges with an applied negative substrate bias Vs of either 0 or 150 V. Oxidation in pure O2 initially occurred at a rate that varied parabolically with time. The oxide overlayers consisted of two partially crystalline sublayers, the upper one Al-rich and the lower one Ti-rich, with no measurable N concentrations in either. Inert-marker transport experiments showed that oxidation proceeded by the simultaneous outward diffusion of Al to the oxide/vapor interface and inward diffusion of O to the oxide/nitride interface. The oxidation rate constant K increased with oxidation temperature Tox at a rate much higher than would be predicted from a simple exponential dependence due to changes in the oxide microstructure (increased crystallinity) with increasing Tox. At Tox ≥850 °C, O transport became the rate-limiting step. After oxidation times, ranging from 6 h at 750 °C to 7 min at 900 °C, oxide crystallites, exhibiting a tetragonal rutile TiO2 structure, were observed in Vs=0 samples to grow at an accelerated rate up through cracks in the oxide overlayer. The formation of these crystallites was postponed until a much later stage in oxide-overlayer development for samples grown with Vs=150 V.