Mark edge jitter model for phase change recording

We model nucleation and grain growth in GeSbTe, and evaluate the jitter based on the randomness of the nucleation process. We consider the case in which an amorphous mark is written on a previously existing amorphous mark. If the previously written amorphous mark was perfectly quenched with no nuclei in it, then we would have all new nuclei followed by their growth. In practice, however, there usually will be some quenched-in nuclei and the number of these will depend on the rate of cooling at the time of formation of the previous amorphous mark. We have studied the effect of these pre-existing nuclei and control of the temperature profile by changing the reflection layer thickness on nucleation, grain growth and jitter.