Abstract
The reaction between a metal and oxygen may be considered to take place in a series of sequential steps, one of which, being much slower than the rest, may be said to determine the reaction rate. In order to determine this rate controlling step and the oxidation mechanism, it is essential to carry out experiments in which the oxidation kinetics are altered in response to a well-defined change in conditions. Having the above point in mind, the effect of external electric currents on protective oxidation (<723°K) of copper and nickel, were used as a useful tool to investigate the oxidation mechanism and particularly the validity of Uhlig's theory of the protective oxidation of metals. The external electron current was supplied by an emitting filament in a triode value arrangement, the plate of which was the oxidizing specimen. This was in turn suspended from a microbalance hangdown, which enabled the oxidation kinetics to be followed directly by a chart recorder. The response of the oxidation kinetics was independent of the applied current, but was related to the presence of the heated filament. This evidence,accompanied by experiments in which various shields were used to mask the specimen from the line-of-sight coupling with the filament, showed that the rate controlling step involved molecular oxygen species. Thus the rate of oxidation is not determined by processes at the metal/oxide interface as in Uhlig's theory, but at the oxide/gas interface. A model is given which leads to the definition of the first electron transfer to molecular oxygen as the rate controlling step in the reaction sequence. It is also shown mathematically that the kinetic data best fits a single stage direct logarithmic rate law rather than the two-stage one proposed by Uhlig. Most of the discussion of logarithmic oxidation has been concerned with the pure metals. Alloys, however, also show similar kinetics of oxidation and the techniques applied to copper and nickel were therefore also applied to an alloy of the two metals. There was very little response to the presence of the hot filament and it was concluded that the oxidation rate is controlled by a different step in this case. The outer surface of thick oxides formed on high copper, cupronickel alloys is invariably copper oxide, yet at room temperature the oxide film is entirely nickel oxide. The ESCA (or XPS) technique was used to determine the temperature at which overgrowth of the initial nickel oxide by cuprous oxide took place on a 60/40 alloy. By means of interrupted oxidation runs on a heated probe in the ESCA spectrometer, the physical and chemical displacement of CU[2]O by NiO was followed. It was concluded that diffusion across the inner NiO phase is probably the rate controlling step during both oxidation and annealing of the alloy at temperatures as low as 433°K, and that the logarithmic law for this alloy is explicable by a modification of Evans' model. The XPS spectra have been interpreted quantitatively and excellent agreement is found between the calculated stoichiometry and the chemical shifts of the Auger spectra. The development of a new technique for conductivity measurements on thin oxide films is also described and is accompanied by results obtained on thin (>10[4]A) films of copper and nickel oxides. It is concluded that this technique may be useful in following the build-up of electron barriers within the oxide.