Theory of Nucleation and Glass Formation

Nucleation is the first step in most first-order phase transitions, which include processes such as gas condensation, solidification, and the fabrication of glass ceramics. It also plays a key role in some biological processes and is fundamentally important in the pharmaceutical industry. Nucleation is commonly described within the framework of the classical nucleation theory, which was first developed to describe gas condensation almost 150 years ago. The classical theory is often called the liquid drop model, assuming the formation of small clusters of the new phase that have a sharp boundary with the original phase. It assumes a single nucleation pathway with clusters growing and shrinking by the addition or subtraction of single molecular units and adopts interface-limited kinetics. However, it is now becoming increasingly clear that nucleation is much more complicated than this. Experiments and computer modeling have demonstrated that the cluster interface is broad, not sharp. Studies of nucleation in colloidal and protein liquids, as well as in silicate glasses, have shown that there can be multiple nucleation pathways, not only one. Nucleation can couple with other phase transitions and ordering processes. The local atomic and chemical order in the original phase, even if it is a liquid or a glass, can couple to the nucleation barrier and may play a role in glass formation in some cases. Finally, studies of nucleation in liquids under a microgravity environment have revealed the important role that stirring can have in nucleation. These points will be briefly covered in this chapter, highlighting experimental and computer modeling results and discussing some new models for nucleation that take account of the diffuse interface and cases where long-range diffusion becomes comparable with the interfacial kinetics.