Why Does Honey Turn to Glass Without Freezing? Eindhoven Physicists Tracked Millions of Simulated Particles and Found the Answer: The Clusters of Molecules That Move Together Grow as a Liquid Cools, Then Shrink Again Below a Certain Temperature, Putting a Brake on Their Own Growth.

A drinking glass is, in a strict physical sense, a liquid that has stopped flowing. It has no crystal lattice, no orderly grid of atoms like ice or salt; its molecules sit in the same disordered jumble they had when it was molten. Yet it is rigid. How a liquid can become a solid without ever changing its structure is one of the oldest unsolved problems in condensed-matter physics, and a doctoral thesis from Eindhoven University of Technology offers a piece of the answer: the molecules that move together in a cooling liquid limit their own growth.

Theoretical physicist Corentin Laudicina, working in the university's Soft Matter and Biological Physics group, studied what happens to viscosity as a liquid is cooled past the point where it would normally crystallize. "As the temperature drops, the molecules become progressively less able to move," he said. "Around the glass transition, however, the viscosity increases extremely rapidly, much faster than you would expect. And yet the material's internal structure barely changes. That is the core of the mystery of glass." The extreme case is the famous pitch drop experiment at the University of Queensland, where a funnel of pitch, rock-hard to the touch, releases a single drop roughly once a decade.

No microscope can follow millions of molecules for the long stretches of time needed to watch this happen, so Laudicina built a "model liquid" of perfectly spherical particles interacting in simple ways and ran it on the TU/e Supercomputing Center. The simulations confirmed something physicists had suspected: a supercooled liquid does not slow down uniformly. Some particles barely move while others form groups that shuffle collectively, "a bit like at a festival," Laudicina said, "where, in the crowded areas, everyone has to move a little if you want to get a beer at the back of the field."

The surprise came when he measured those clusters carefully and tracked them over time as the temperature fell. At first the clusters grew, as expected. But below a certain temperature they began to shrink again. Older theories, which do not account for the clusters at all, predicted they would keep growing without limit until the liquid abruptly seized up. The simulations showed the opposite: the clusters act as a natural brake on themselves. That, Laudicina argues, is why no sharp, sudden arrest is ever seen in the laboratory or in simulations, only a smooth, if steep, climb in viscosity.

"Our equations were missing a kind of feedback mechanism," he said. "They could predict that the clusters would grow, but not that their growth would eventually slow itself down. Once we incorporated that feedback into the theory, and that was certainly not an easy task, the sharp transition previously predicted by the theory disappeared. What remained almost perfectly matched what we see in the simulations." The thesis, "Simple Fluctuations in Simple Liquids," was published by Eindhoven University of Technology under DOI 10.6100/mwnc-0456.

Laudicina is careful to call the work fundamental rather than applied, but he points out that the same mathematics keeps showing up in places that have nothing to do with window panes. Cells in a tumor are packed together like the particles in his simulations; they can jam in the same way, and then suddenly loosen and invade surrounding tissue. His group is working with cell biologists on the idea that this is, physically, a glass transition. Related problems appear in computer science and machine learning, where systems of many interacting parts get stuck in rugged landscapes of possible states.

"That is why you have to be careful about labeling this kind of research simply as 'knowledge for knowledge's sake,'" he said. "Precisely because the same ideas keep appearing in places where you would not expect them, it is certainly worth understanding them at a fundamental level." The glass on your table has been keeping this secret for about five thousand years. The claim from Eindhoven is that the molecules inside it never froze; they simply crowded each other until the crowd could no longer grow.