The oxidation of tungsten and molybdenum occurs by two consecutive reactions, forming first a suboxide then the trioxide. Tantalum and columbium oxidize by four simultaneous reactions: solution of oxygen in the metal, nucleation and growth of a suboxide phase at the metal surface and two phase boundary processes giving rise to two different modifications of the pentoxide. By assuming that all reactions are first order complex chain reactions, rate equations are formulated giving the rate of oxidation as a function of pressure, temperature and time. Regression rate expressions for the metals tungsten, tantalum and columbium above 700 C are given as;; and, respectively. The rate is expressed in cm/hr, T is in degrees K and Po2 in atmospheres pressure of oxygen. 3.8.4, 2.1,1, 6.3.5, 6.3.16, 6.3.13

The oxidation of Mond nickel was studied between 1000 ~ and 1200~ at oxygen pressures from 6.5 x 10 -3 to 20.4 arm. The oxidation rate was controlled by diffusion through the oxide and had a gross activation energy of 50 _ 4 kcal/mole. Least squares straight line fitting over the entire pressure range indicated the rate was proportional to Po21/5. However, the rates above 1 arm appeared insensitive to pressure. On this basis, the measured rates below 1 atm more closely follow the PO21/4 dependence expected from electrical conductivity studies in NiO. The oxide oriented in a preferred manner with 100 planes parallel to the polycrystalline nickel base.In the early work of Wagner and Griinewald (1) on the oxidation of nickel, the observed parabolic rate was derived on the basis of diffusion of nickel ions by a vacancy exchange mechanism through a uniform layer of NiO. Below 1000~ the parabolic relation has been observed by other investigators (2-4). The role of cation vacancies has been confirmed by electrical and thermoelectric studies on NiO (5-7), which is a p-type semiconductor, and by the close correspondence of the activation energies for oxidation and for diffusion of radioactive nickel in NiO (8,9). Nevertheless, there are some anomalies. Gulbransen and Andrew (10) have observed that at 900~ and higher, parabolic rate law plots of the data show increasing values of the rate constant with time. The dissociation of two electron holes from each cation vacancy and a resulting oxidation rate and electrical conductivity dependence on Poe 1/6 which was assumed by Wagner and Griinewald are not clearly supported by electrical conductivity measurements or oxidation rate measurements.

Kinetic data are reported for the solvolysis in water of propane-2-nitrate, butane-2-nitrate, cyclopentyl nitrate, and cyclohexyl nitrate. In each case, the dependence of rate constant on temperature is analysed in terms of two mechanisms for the solvolytic reaction. First it is assumed that the rate constant describes a single step reaction, the analysis leading to estimates of the heat capacity of activation ΔCp≠. Three different analytical methods are discussed in this regard. Second it is assumed that the rate constant describes a two stage mechanism, the first stage being reversible. In this case the explanation of the ΔCp≠ term calculated according to the first mechanism is quite different. We comment on the alternative explanations of trends in activation parameters.

By regarding the kinetics of oxidation of tungsten as occurring by the consecutive reactions: W + O~ --> WO~; WO~ + %O~ -> WO~(~); WO~(~) 1/n (WO~),~v~p) and by expressing the rate in terms of conventional chemical kinetic quantities, it is shown that the results of three investigations may be correlated in the temperature range 500~176 and pressure range 0.0013-20.8 atm of oxygen. A good approximate form for the rate of attack of metallic tungsten above 700~ is shown to have the form: Rate = d(m/A)/dt -~ 5.89 X 10 ~ exp (--12,170/T) P~/'~ mgw/cm ' hr, where T is expressed in ~ and P in atmospheres of oxygen.

No abstract