Bulk metallic glass gears in a laboratory test with a gloved hand

· Translated from the Spanish original · 中文 · 日本語

Materials physics

In metallic glasses, local order and density fluctuations are born together as crystals form

A PNAS study revisits the idea that crystals start from already ordered zones and shows density and order emerge at once in disordered systems.

What happened

  • Nine researchers, with Xu, Wang, Cao and Ding among the authors, published on 6 November 2026 in PNAS a study on how crystallization begins in disordered materials.
  • Crystallization occurs in pharmaceuticals, metallurgy, polymers, biomolecules and glass formation, but how it proceeds at its earliest stage remains unclear.
  • Studies on glass-forming liquids suggested that crystals are born in structurally preordered regions, with density fluctuations playing a secondary role. That idea was generalized from hard-sphere colloids.
  • The team shows, in disordered metallic systems, that density fluctuations and structural preordering emerge simultaneously during nucleation.

Why it matters

  • Crystallization is a universal process, and understanding its first step helps control it: to avoid it, as when making metallic glasses, or to encourage it.
  • The result corrects a generalization that came from model systems and tests it on real metals.
  • The authors propose a unified framework of nucleation in disordered materials.

The number

From pure elements to alloys. The behaviour was observed across disordered metallic systems ranging from pure elements and intermetallic compounds to glass-forming alloys, which gives the result generality.

Context

The coupling they describe reflects local atomic rearrangements in which free-volume reduction, denser packing and bond-orientational order develop together. They are not processes that occur one after the other.

A metallic glass is an alloy with no crystalline structure, and making one requires cooling so fast that atoms cannot arrange themselves. Knowing how order begins helps understand how much room there is to prevent it.

The abstract does not explain the analysis methods or how many systems were studied, and this note is based only on it. That is why the result should be read as a correction to a widespread view, and not as a definitive theory of nucleation.

At Mamífero we follow the physics of materials. We reported that copper oxide grows outward and inward at once and that a one-metre titanium single crystal was 3D-printed. Also that PolyU presented an alloy that keeps strength and ductility at cryogenic temperatures and that Columbia trapped electron waves in a metal by interference.

What’s next

  • Other groups can test whether the coupling holds in non-metallic disordered materials.
  • Crystallization models can incorporate the simultaneous emergence of order and density.
  • A finer understanding of nucleation could help design more stable metallic glasses.

Bottom line

A crystal does not seem to be born first one way and then another. Order and density appear together, and that changes how the first step is thought of.

Sources

  • Xu, Wang, Cao, Ding et al., “Coupling of structural preordering and density fluctuation: A unified picture of nucleation in metallic glasses”, PNAS, 6 November 2026, DOI 10.1073/pnas.2613542123. This note is based on the paper’s abstract.
  • Cover image: NASA Kennedy Space Center / NASA/Kim Shiflett, Public domain, via Wikimedia Commons

Edited by Rodrigo Cornejo. How we select and verify: who writes these notes.

Related notes

← All notes