The defining phase transformation associated with the glassy state is the glass transition. However, this is not a thermodynamic phase transition at all, but rather a kinetic transformation as a liquid is supercooled. The glass transition involves a freezing of the configurational degrees of freedom at the nanoscale level such that the glass is a solid-like material but with liquid-like properties. In recent years, much progress has been made in understanding the fundamental physics of the glass transition.
The glass transition itself is a function of both the structural relaxation time — an intrinsic property —and an external observation time scale (i.e., the relevant time scale of the experiment). Glass transition can be induced through either of two pathways. The first is by increasing the internal relaxation time of a glass forming liquid. This is what happens in the traditional approach to glass-forming by cooling from a high-temperature melt. As the temperature of the system is lowered, the atoms lose thermal energy, eventually becoming trapped in a particular configuration. The second path to induce glass transition is by shortening the observation time scale. Any liquid, when observed on a short enough time scale, will behave as a solid material, that is, a glass. This method of inducing glass transition is seen, for example, in specific heat spectroscopy experiments where a system is probed on a multitude of disparate time scales.
The glass transition is fundamentally kinetic in origin, but it has important thermodynamic consequences. While the freezing of the configurational degrees of freedom during the glass transition does not result in any change in volume or energy, the configurational contributions to entropy and second-order thermodynamic properties such as heat capacity and thermal expansion coefficient are all lost as the system becomes frozen at the nanoscale level. This has led many scientists to postulate the existence of an “ideal” glass transition that entails a second-order thermodynamic phase transformation. However, the concept of an ideal glass transition has been widely criticized based on both theoretical and experimental considerations.
While the glass transition involves the imposition of a kinetic constraint, the opposite process — structural relaxation — involves the lifting of this constraint. Because glass is a nonequilibrium material, it is continually relaxing to approach its metastable supercooled liquid state. At low temperatures and short time scales, this relaxation process is effectively frozen. However, at higher temperature or for exponentially longer times,
measurable relaxation of the glassy state can be observed, reactivating the nanoscale transitions that ultimately lead to viscous flow. Relaxation is a spontaneous process and involves an increase in heat capacity, thermal expansion coefficient, and entropy as the configurational degrees of freedom of the system become activated.