The Thermodynamic and Structural Origins of the Glass Transition

You are reading version 1 of this direction, frozen 2026-10-04. The latest version is version 2.

Version 1 of 257 papers77 conceptsfrozen 2026-10-04

When a liquid cools into a crystal, the change is obvious: atoms snap into a neat, repeating grid. But when a liquid cools into a glass, it presents a profound paradox. Its viscosity increases by fifteen orders of magnitude—turning from a runny fluid into a rigid solid—yet its atomic structure remains almost as disordered as the liquid it came from.

For decades, researchers have asked: Is the glass transition a purely kinetic "traffic jam" where particles simply lose the energy to move, or is it a true thermodynamic phase transition driven by a hidden structural order?

To understand why a liquid stops flowing, we have to look at its "Potential Energy Landscape" (PEL). Imagine a high-dimensional map where the "elevation" represents the energy of a specific arrangement of all atoms in the system.

At high temperatures, a liquid has enough thermal energy to sail over the peaks of this landscape. But as it cools, it begins to sink into the valleys. Francesco Sciortino and P. Tartaglia showed in The phase space structure of supercooled liquids that there is a clear "topographical" signature to this slowing down.

As the liquid enters the "supercooled" regime, its dynamics change from a continuous flow to a series of discrete "hops" between stable configurations. Thomas B. Schrøder and Sharon C. Glotzer provided numerical evidence in Crossover to potential energy landscape dominated dynamics that this transition occurs when the system becomes "trapped" in local energy minima, or "inherent structures."

The Entropy Crisis and the Adam-Gibbs Relation

If the glass transition is thermodynamic, there must be a measurable property that signals the change. The leading candidate is configurational entropy ($S_c$)—a measure of the number of distinct "valleys" or arrangements available to the liquid.

In A first principle computation of the low temperature thermodynamics of glasses, Giorgio Parisi used the mathematics of spin glasses to bridge the gap between microscopic potentials and macroscopic thermodynamics. The idea, popularized by the Adam-Gibbs relation, is that as a liquid cools, the number of available states ($S_c$) drops. To move, particles must rearrange cooperatively in increasingly large groups.

Edmund A. Di Marzio argued in Configurational entropy approach to the kinetics of glasses that this loss of entropy is the fundamental driver of viscosity. When $S_c$ vanishes at a theoretical "Kauzmann temperature," the liquid would reach an "ideal glass" state—a single, perfectly optimized disordered arrangement.

Finding the Hidden Length Scale

If particles must move in "cooperative groups," there should be a physical size to these groups. However, standard structural tests (like two-point correlation functions) show no growing "order" as a glass forms.

To find this hidden scale, researchers developed the Point-to-Set method. Imagine "pinning" all particles in a liquid outside of a sphere and seeing how the particles inside the sphere respond. Guilhem Semerjian proved in Rigorous Inequalities Between Length and Time Scales in Glassy Systems that the relaxation time is mathematically bound to the growth of this spatial correlation length.

This supports the Random First Order Transition (RFOT) theory. Giulio Biroli and Chiara Cammarota describe the glass as a "mosaic" of different local arrangements. In Confinement as a Tool to Probe Amorphous Order, they showed that by squeezing a liquid into small spaces, we can finally "see" the amorphous order that governs the transition.

Local Barriers vs. Cooperative Motion

A major tension in the field is whether this "mosaic" of cooperative motion is actually necessary. Some researchers, like Michael E. Cates in Does a Growing Static Length Scale Control the Glass Transition?, argue that local kinetic barriers—the simple difficulty of a particle squeezing past its neighbors—account for most of the slowdown.

Similarly, Massimo Pica Ciamarra used the SEER algorithm in Local vs. Cooperative to suggest that the rising cost of small-scale, local movements is the primary driver, rather than massive group coordination.

New Frontiers: Machine Learning and Atomic Tomography

Recent work has begun to resolve these conflicts by finding structural signatures that were previously invisible to the human eye.

  1. Machine Learning "Softness": Andrea J. Liu and Olivier Dauchot demonstrated in Attractive vs. truncated repulsive supercooled liquids that while average structures look identical, subtle fluctuations in the "pair correlation function" can predict which particles are "soft" (likely to move) and which are "hard" (jammed).
  2. Direct Observation: For the first time, Yakun Yuan and colleagues used atomic electron tomography in Direct observation of 3D atomic packing to map the 3D positions of atoms in an amorphous material. They found "pentagonal bipyramid networks" that provide a medium-range order previously only theorized.

By combining the abstract thermodynamics of Parisi with high-resolution structural analysis, the field is moving toward a unified view: the glass transition is not just a traffic jam, but a complex dance where the geometry of the energy landscape and the loss of available states force atoms into a "fuzzy mosaic" of arrested motion.

Papers in this version

History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 27 papers
  2. Version 1 · 2026-10-04 · 57 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_structural_dynamic_decoupling_5c27422f,
  title = {structural-dynamic decoupling},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/structural-dynamic-decoupling/rdr_5c27422fd9c2425b824ae1ea1c3a8301},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_the_thermodynamic_and_structural_origins_5c27422f_v1,
  title = {The Thermodynamic and Structural Origins of the Glass Transition},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/the-thermodynamic-and-structural-origins-of-the-glass-transition/rdr_5c27422fd9c2425b824ae1ea1c3a8301/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}