Dependence of atomic diffusion in GaAs upon experimental design

The rate of impurity diffusion in GaAs depends upon the concentration of point defects such as vacancies and interstitials. The concentration of those point defects is strongly affected by the region of the phase diagram corresponding to the chosen experimental design. Different regions of phase diagrams are used to illustrate why nominally similar experiments can lead to dramatically different results. When the equilibrium point defect concentrations can be approached in a time which is short compared to the anneal time, the measured diffusivities tend to be reproducible. However, when the point defect concentrations approach their equilibrium values over times which are long compared to the anneal time, the measured diffusivities tend to be irreproducible and time dependent. Examples of "anomalous" diffusion are discussed and shown to be associated with fundamental differences in experimental design that tend to bring the point defect concentration toward equilibrium in different regions of the phase diagram.