Oxide thickness growth

The purpose of this review paper is to survey the principles of high temperature oxidation or high temperature corrosion. A typical situation is that of a metal exposed to a hot gas which can act as an oxidant. In many cases the oxidation product forms a layer which separates the reactants, the metal and the gas atmosphere. Under special conditions, the kinetics are diffusion controlled, i. e,, the rate of the reaction (the rate of oxide thickness growth) depends on the diffusion of species, ions and electrons, through the layer (sometimes called a tarnish layer). Actually when a metal or alloy is exposed to a corrosive gas, the reaction kinetics may be controlled by one or more of the following steps ... 

The mathematics of oxidation kinetics involves diffusion and migration mass transfer. According to Pick s first law of diffusion, the molar flux is related to the rate of oxide thickness growth (dx/dt) is given by... 

In fact, X in eq. (10.19) represents an external thickness in most cases. Now, the generalized equation for die rate of oxide thickness growth takes the form... 

The service temperature of a particular stainless steel can be found elsewhere [17]. Recently published stainless steel, carbon steel and nickel data on oxidation clearly exhibit increasing oxide thickness growth with increasing temperature as summarized in Figure 10.13 after one year of exposure in air [18]. 

A wide variety of in situ techniques are available for the study of anodic hhns. These include reflectance, eUipsometry, X-ray reflectivity, and SXRD. X-ray reflectivity can be used to study thick surface layers up to 1000 A. The reflectance technique has been used to study oxide growth on metals, and it yields information on oxide thickness, roughness, and stoichiometry. It the only technique that can give information on buried metal-oxide interfaces. It is also possible to get information on duplex or multiple-layer oxide hhns or oxide hhns consisting of layers with different porosity. Films with thicknesses of anywhere from 10 to 1000 A can be studied. XAS can be used to study the chemistry of dilute components such as Cr in passive oxide hhns. 

The second current maximum J3 corresponds to an oxide thickness at which tunneling of charge carriers becomes negligible, as shown in Fig. 4.7. At the bias corresponding to J3 the formation of anodic oxides in electrolyte-free HF shows a change of growth kinetics, as shown in Fig. 5.2. 

The growth rates of anodic oxides depend on electrolyte composition and anodization conditions. The oxide thickness is reported to increase linearly with the applied bias at a rate of 0.5-0.6 nm V-1 for current densities in excess of 1 mA cnT2 and ethylene glycol-based electrolytes of a low water content [Da2, Ja2, Crl, Mel2] (for D in nm and V in V) ... 

For a finite flux jA, there is a (steady state) shift of the AX crystal towards the side with the higher pXi. jx does not lead to such a shift. The shift velocity is / A-Vm(AX). Equation (4.104) can also be used to quantify the basic (one dimensional) metal oxidation experiment A+1/2X2 = AX shown in Figure4-4. In terms of thickness growth, one obtains from Eqn. (4.104) the expression... 

Then the surface of the undoped layer must be oxidized. This can be done in two ways (1) The sample can be exposed to air at room temperature for several days to permit the growth of a thin oxide layer and then the oxide thickness can be evaluated according to available data on the oxidation of a-Si H as a function of time, already examined by Ponpon and Bourdon (1982). (2) The oxidation could be obtained by leaving the sample (Abeles et al., 1981 Wronski et al., 1981), at a given temperature (100- 200°C) in a cleaned chamber, in the presence of pure Oz. Finally, a Pd film, whose thickness can range between 80 and 500 A, can be thermally evaporated to complete the procedure. Photolithographic techniques and lift-off procedures can be used to define the catalytic area. Figure 6 shows a schematic of a typical MIS structure with the overall dimensions. 

In an earlier study we had reported the XPS analysis of tungsten oxides formed during anodic polarization experiments. It was determined that even at high applied potentials, the oxide thickness values are less than the mean free path of electrons in the oxides (generally assumed to be between 30 to 50 A ). Clearly the oxide growth in tungsten is a slow process. However, despite the relatively small thickness vsilues, the steady state current density during anodic polarization is restricted to a few tens of microamperes. 

---