No abstract

A method is proposed to allow a more accurate evaluation of thermogravimetric data to identify diffusion or partial diffusion control of scaling kinetics. This method is based on the ® tting of mass-gain data to a parabola over a short time interval. The translation of the time interval over the entire test time period provides an actual instantaneous parabolic rate constant independently of any transient stage or simultaneous reaction steps. The usefulness and limitations of this procedure are illustrated from oxidation tests performed on several metallic materials (pure nickel, single-crystal superalloys, and Nb± Ti± Al alloy).

The oxidation kinetics of high-purity nickel were studied between 500 and 1200°C, in pure oxygen at atmospheric pressure, for aûerage oxide-scale thicknesses of 1, 5, 10, and 30 µm. In the oûerall temperature range studied, a decrease in the parabolic rate constant k p with increasing scale thickness was obserûed. Depending on temperature and oxide-scale thickness, growth kinetics can be interpreted as a mixture of parabolic-and cubic-growth kinetics. Possible correlations between growth kinetics and microstructures of the oxide scales were inûestigated. From this set of experimental data, oxidation-kinetics models were tested. In particular, the effect of grain-boundary diffusion on NiO-growth kinetics was discussed. The correlations between growth kinetics and oxide microstructures appear to be more complex than usually reported.