Researchers Observe Motionless Atoms Inside Fully Molten Metal

Professor Andrei Khlobystov at the University of Nottingham and his collaborators have identified an unusual atomic state inside molten metal nanoparticles that challenges conventional descriptions of liquids and solids. Working with researchers at the University of Ulm in Germany, the team observed that even when metal particles are fully molten, certain atoms can remain fixed in position. Their findings, published in ACS Nano, suggest that localized atomic immobility within a liquid can significantly alter how metals freeze and may give rise to a confined, nonclassical phase.

Leist, C., Ghaderzadeh, S., Kohlrausch, E. C., Biskupek, J., Norman, L. T., Popov, I., Alves Fernandes, J., Kaiser, U., Besley, E., & Khlobystov, A. N. (2025). Stationary Atoms in Liquid Metals and Their Role in Solidification Mechanisms. ACS Nano, 19(50), 42002–42012. https://doi.org/10.1021/acsnano.5c08201

The study addresses a long-standing challenge in materials science: understanding how liquids transition into solids at the atomic level. In simple terms, solids are characterized by atoms arranged in ordered positions, while liquids are defined by continuous atomic motion. Yet this distinction becomes less clear at the nanoscale, where surfaces, defects and confinement can influence atomic dynamics in ways not captured by bulk thermodynamic models.

Professor Andrei Khlobystov at the University of Nottingham stated,

“The discovery of a new hybrid state of metal is significant. Since platinum on carbon is one of the most widely used catalysts globally, finding a confined liquid state with non-classical phase behaviour could change our understanding of how catalysts work. This advancement may lead to the design of self-cleaning catalysts with improved activity and longevity.”

To investigate solidification in real time, the researchers used advanced transmission electron microscopy to observe metal nanoparticles as they melted and cooled. The experiments were performed at the SALVE microscopy center at Ulm University, where Dr. Christopher Leist conducted imaging using a low voltage instrument capable of resolving individual atoms while minimizing beam damage. The nanoparticles, composed of platinum, gold and palladium, were deposited on atomically thin graphene sheets that served as both support and heating platform.

As expected, heating the particles above their melting points led to rapid atomic motion consistent with a liquid state. However, detailed analysis revealed that not all atoms behaved the same way. A subset of atoms remained effectively stationary, even at temperatures where full mobility would typically be assumed. These atoms were found to be strongly bound to defect sites in the graphene support.

Further experimentation showed that the number of immobile atoms could be influenced by the electron beam itself. By focusing the beam, the researchers were able to generate additional defects in the graphene, which in turn anchored more metal atoms in place. This provided a means of controlling atomic pinning within the liquid droplet.

The presence of stationary atoms had a pronounced effect on how the molten nanoparticles solidified. When only a few atoms were pinned, the particles froze in a conventional manner. Crystallization began at nucleation sites and spread until a well ordered solid structure formed.

When the density of pinned atoms increased, the freezing pathway changed. If immobile atoms were distributed around the perimeter of the droplet, they formed what the researchers describe as an atomic corral. Within this confined boundary, the interior atoms remained mobile but were prevented from initiating the usual crystallization process. As a result, the liquid state persisted at temperatures significantly below the normal freezing point.

In the case of platinum, the confined liquid remained stable at temperatures hundreds of degrees lower than expected for bulk material. Eventually, further cooling led not to a crystalline solid but to an amorphous metallic phase lacking long range atomic order. This state was metastable and dependent on confinement. When the pinned atoms were disrupted, the accumulated strain was released and the material reorganized into its standard crystalline configuration.

The team complemented their microscopy observations with theoretical modeling. Simulations explored how reduced atomic mobility at the droplet boundary affects nucleation and growth. The results indicated that immobilized atoms can suppress the formation of critical nuclei required for crystallization. In effect, the pinned atoms create a barrier that delays or blocks the transition from liquid to crystal.

This finding adds nuance to classical descriptions of phase transitions. Traditional thermodynamic models assume that once a material exceeds its melting point, atomic mobility becomes uniform. The new evidence suggests that at the nanoscale, surface interactions and local defects can produce heterogeneous mobility within a liquid. These localized constraints can stabilize otherwise unstable states.

The implications extend beyond fundamental phase theory. Platinum supported on carbon materials is widely used in catalytic systems, including fuel cells and chemical synthesis. Catalytic performance depends sensitively on atomic arrangement and surface structure. If confined liquid like states can form under operating conditions, they may influence reaction pathways, stability and durability.

Dr. Jesum Alves Fernandes, who contributed expertise in catalysis, notes that understanding these hybrid states could inform the design of catalysts that maintain activity over extended periods. If atomic pinning can be controlled deliberately, it may be possible to engineer surfaces that resist degradation or self reorganize under reaction conditions.

The research also builds on earlier work by the Nottingham and Ulm teams in atomic scale imaging. In previous studies, they recorded chemical bond formation and breaking in real time using similar microscopy techniques. The current investigation applies those capabilities to phase transitions rather than chemical reactions, but the underlying approach is the same: direct observation of matter one atom at a time.

The concept of corralling has previously been demonstrated in electronic and photonic systems, where electrons or photons are confined to defined regions. In this case, the confinement involves atoms themselves. The study represents one of the first demonstrations that atomic mobility within a liquid can be locally restricted in a controlled manner, producing hybrid behavior that blends characteristics of both solid and liquid phases.

Looking ahead, the researchers aim to explore methods for deliberately positioning pinned atoms rather than relying on incidental defects. If atomic corrals can be designed with precision, it may become possible to stabilize novel phases or manipulate solidification pathways in a predictable way. Such control could be relevant not only for catalysis but also for additive manufacturing, alloy design and nanostructured materials.

The work was supported by the Engineering and Physical Sciences Research Council through a programme focused on metal atoms at surfaces and interfaces. While the experiments were conducted under specialized laboratory conditions, the principles revealed by the study are broadly applicable. They suggest that at sufficiently small scales, the boundary between liquid and solid is less absolute than once assumed.

Rather than a simple transition from motion to order, the freezing of a nanoscale metal particle can involve confined liquid regions, pinned boundary atoms and metastable amorphous states. For materials scientists and engineers, this discovery highlights the importance of local structure and confinement in governing phase behavior. As experimental techniques continue to improve, additional hidden states may emerge in systems long thought to be well understood.