Kinetic Decomposition of Compounds in Chemical Potential, Gradients

Finally, we mention that the (Mg,Co)0 crystal surface with the lower oxygen potential is morphologically unstable. Therefore, in a strict sense the boundary conditions (which have been tacitly assumed to be stationary) are not time-independent. This phenomenon will be discussed further in Chapter 11. 

In concluding, let us comment on the time needed to attain the steady state after establishing the surface activities. Two transient processes having different relaxation times occur I) the steady state vacancy concentration profile builds up and 2) the component demixing profile builds up until eventually the system becomes truly stationary. Even if the vacancies have attained a (quasi-) steady state, their drift flux is not stationary until the demixing profile has also reached its steady state. This time dependence of the vacancy drift is responsible for the difficulties that arise when the transient transport problem must be solved explicitly, see, for example, [G. Petot-Er-vas, et al. (1992)]. 

If we could arrange the demixing experiment (Fig. 8-2) such that the vacancy flux (caused by the activity difference at opposite surfaces) remains constant and the crystal therefore shifts with constant velocity, we could calculate the time required to attain the steady state with 

Kinetic demixing stems from activity differences established between the opposite surfaces. These differences can be produced in various ways. In this section, we applied buffers with different chemical potentials. Other possibilities are activity changes through temperature gradients and activity changes through stress gradients. These situations will be discussed in Sections 8.5 and 8.6. 

3 Kinetic Decomposition of Compounds in Chemical Potential Gradients 

---