Skip to main content

Morning Overview

Physicists filmed magnetic whirlpools melting out of their lattice for the first time

Under a magneto-optical Kerr microscope, a regular array of magnetic skyrmions, each one a tiny swirl of magnetisation, starts to lose its pattern. First the gaps between neighbouring swirls stop being even. Later the rows no longer point the same way, and the lattice dissolves into a loose scatter. Physicists at Johannes Gutenberg University Mainz recorded that sequence in real time in a thin magnetic film, and the footage underlies a Nature Nanotechnology paper published on 4 August 2025.

Stay ahead on AI: Join 600,000+ professionals who read The Deep View, the daily newsletter that keeps you up to date with artificial intelligence.

Subscribe to The Deep View →

We may earn a commission if you sign up.

The detail that sets the experiment apart is what caused the collapse. Heat did not melt the lattice. The researchers shrank the skyrmions by adjusting the magnetic field, and the shrinking alone was enough to set them loose.

Two steps from order to disorder

In the Mainz release of 4 August 2025, the group describes a lattice that gives up its order in two stages. Translational order goes first: the skyrmions stay inside a lattice-like arrangement, but the distances to their nearest neighbours turn irregular. Orientational order goes second, and with it the lattice dissolves completely. The release contrasts the two-dimensional case with three-dimensional lattices such as ice, which do not come apart in this distinct two-step fashion.

Raphael Gruber, the paper’s first author, framed the goal as a simple question: what happens when an ordered state is reverted to a disordered one. The university’s own description of the answer is “the first direct, real-time observation at the microscopic level of a 2D ordered lattice becoming disordered.” Gruber said the skyrmions, which he called “miniature magnetic vortices,” made it possible “to directly observe, for the first time,” how that transition unfolds.

An earlier version of the work had been posted to arXiv on 22 January 2025 as a preprint titled “Imaging Topological Defect Dynamics Mediating 2D Skyrmion Lattice Melting”, with Gruber, Kläui and Sudbø among its authors alongside Peter Virnau and others. Its abstract spells out what the two steps mean: melting passes through an intermediate hexatic regime between the solid lattice and the isotropic liquid. In that middle state the skyrmions have lost their fixed positions but still share a common orientation, and only then does the orientation go too.

The preprint adds a measurement that the news coverage skips. Skyrmion size and an effective temperature are both tuned during the experiment, which lets the team drive the melting, and the authors quantify how often topological defects occur and track the motion of lattice dislocations, defects in the pattern that were previously inaccessible to direct view. The dislocations’ diffusion coefficient, they report, is orders of magnitude higher than that of the skyrmions themselves, so the flaws in the pattern travel far faster than the swirls that make it up.

Shrinking the whirlpools instead of heating them

Skyrmions are generated by carefully calibrating temperature and magnetic field, and packed densely enough they organise themselves into a regular lattice. Raising the temperature to melt that lattice would have altered the very conditions that create the vortices, so Gruber and colleagues went another way. They reduced the size of the skyrmions “by modulating the magnetic field,” which, as Gruber put it, gave the swirls greater mobility within the lattice and let them move. The microscope does the filming. A Kerr microscope reads magnetisation from how a surface changes the light reflected off it, so each skyrmion shows up as a spot whose position can be followed from frame to frame.

Mathias Kläui, whose group at Mainz carried out the work, credited collaboration with colleagues for making the melting transition easier to untangle, and described topology, together with the dynamics of topological properties, as “a central research focus” for many scientists in Mainz. The full account is in the Nature Nanotechnology paper, by Gruber, Röthörl, Fröhlich and colleagues, volume 20, issue 10, page 1405, titled “Real-time observation of topological defect dynamics mediating two-dimensional skyrmion lattice melting.” Funding came from the ERC Synergy Grant 3D MAGiC and the TopDyn research initiative of Rhineland-Palatinate.

The computing case, and its limits

A second institution joined the story in a piece dated 24 February 2026 from the Norwegian University of Science and Technology. NTNU’s Asle Sudbø, who heads the Centre for Quantum Spintronics and took part in the work, put the experiment plainly: “For the first time, researchers have managed to film how a magnetic skyrmion lattice melts in real time.”

Sudbø went further on what skyrmions might become. He described them as small magnetic vortices that can store information in a completely new way, and said nanoscale vortices of this kind “could help develop computers that resemble biological brains.” He contrasted that with current technology, which keeps demanding more space and more energy to process data, and noted that the shape and size of individual skyrmions can be changed on the fly while they are being filmed.

Every one of those computing claims is prospective. The experiment studied the physics of a lattice in a magnetic film, and no device was built or tested. The aggregator ScienceDaily carried the story again on 26 October 2026, more than a year after the paper, with the same stated implication: smaller, more energy-efficient brain-like computing, as a possibility. What the footage settles is a narrow question about order and disorder in two dimensions. Whether the computing ambitions follow is left open, and Sudbø’s own summary is modest on that point: the findings, he said, give insight into how these types of particles behave.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

Stay ahead on AI: Subscribe to The Deep View →

We may earn a commission if you sign up.


More from Morning Overview

Morning Overview is reader-supported. Some links in our articles are affiliate links, and we may earn a commission at no extra cost to you. As an Amazon Associate I earn from qualifying purchases. Full disclosure.