Void migration theory

Based on the observation of numerous porosities in the scales, the authors developed the famous void migration theory to explain the increased oxidation rates in the presence of water vapour and carbon dioxide at 850°C and 950°C. While the void migration theory was readily accepted in the high temperature corrosion community [39], most of the porosities presented in the Rahmel andTobolski study were more likely caused by poor metallographic preparation. Porosities in scale will be further discussed in this chapter. 

The void migration theory proposed by Rahmel and Tobolski [38] can be used to explain the improved scale adhesion and hence increased oxidation rate of steel as a result of H2O presence in air or oxygen when genuine... 

In summary, it has been observed that the presence of water vapour in air or oxygen improves the scale adhesion formed at high temperatures. When voids or pores are generated in the scale, the void migration theory proposed by Rahmel and Tobolsk [38] can be used to explain the improved scale adhesion. However, when voids or pores are not present in the scale, the void migration mechanism is invalid. Two other promising theories have been proposed to explain the improved scale-steel adhesion in this latter case. One is the traditional theory of improved plasticity as a result of improved dislocation mobility in the wustite scale due to the presence of hydrogen in certain forms in the scale. The other is the balanced transport theory recently developed by Hultquist et al. However, the actual mechanisms by which these proposed theories work are still not clear. 

Migration and coalescence of voids as weU as their interactions with grain boundaries (GBs) in the presence of the electric wind force is crucial for understanding the failure mechanism. The first fundamental theory of void migration was developed by Krivoglaz [34] for an isolated spherical void and was later modified by Ho [35] for voids in the vicinity of an external surface. At that, the theory of electron wind force [36, 37] was used to demonstrate a (l/R)-size dependence of void velocity. However, the interaction of a void with GBs during electromigration (EM) was not considered. 

---