A University of Chicago team has measured a charge-ordered state in the two-dimensional van der Waals magnet Fe₅GeTe₂ in which electrons move collectively at low velocity while remaining quantumly coherent. The configuration qualifies as a Kondo-like phase, according to the researchers.

The work, led by Shuolong Yang, appears in Science Advances. It marks the first experimental evidence that an interaction-driven flat band can drive electronic ordering through flat-band nesting — without Moiré stacking or geometrically frustrated engineering.

What did the team actually measure?

Using angle-resolved photoemission spectroscopy (ARPES), Yang's group probed individual phase regions of Fe₅GeTe₂ that measured tens of microns across.

The charge order originated from folding of electronic bands in the Brillouin zone within 30 meV of the Fermi level. That narrow window places the relevant physics at the Fermi edge, which governs low-temperature transport.

"We were able to surgically probe the electronic band structure of each individual phase region ... and figure out the physics," Yang told Physics World.

The material hosts multiple structural phases with nearly degenerate energies. Each phase can be stabilized at room temperature, the team found. That structural flexibility raised two questions for Yang: what is the nature of each phase, and can the near-degeneracy encode information?

Why does coherence at slow speeds matter?

Extremely flat electronic bands usually become incoherent once carrier interactions grow too strong. Fe₅GeTe₂ instead produced "quantum coherent flat bands right at the Fermi level," Yang said.

The discovery places an effective Kondo lattice inside a ferromagnetic phase — a configuration theoretical models had not predicted. Most Kondo-lattice systems studied to date sit in non-magnetically-ordered hosts.

"We think that our surprising finding originates from a many-body localized state interacting with the conduction electrons," Yang said.

For R&D managers tracking 2D-materials portfolios, the result reframes flat-band physics. Researchers previously needed to twist bilayer materials or engineer geometric frustration to obtain such states. A stoichiometric compound removes that fabrication complexity from the synthesis chain, though the temperature constraint remains.

The team's ARPES measurement focuses on surface states and on micron-scale phase regions. Whether the coherent flat-band behavior extends uniformly through bulk Fe₅GeTe₂ samples is not yet established in the published data.

How could this connect to memory devices?

The group's follow-on experiment uses micro-focused laser pulses at multiple frequencies to switch Fe₅GeTe₂ between its observed Kondo-like phase and adjacent structural phases.

If switching proves reliable, the many-body physics could underwrite memory operations — the second research question Yang posed after the structural work.

"We're using carefully designed laser pulses of different frequencies to switch between the multiple phases of Fe₅GeTe₂," Yang said. "If successful, we will utilize the many-body physics for some real memory operations."

The laser scheme amounts to a feasibility test for the encoding idea. It does not yet yield a working memory cell, and the team has not disclosed switching speed, endurance, or write-energy figures.

What stands between observation and a usable device?

The quantum effects have so far manifested only at ultralow temperatures. Room-temperature operation is the obvious next gate for any commercial memory application, and it remains unmet.

Yang also flags open questions about topology, structural defects, and a complete microscopic description of the Kondo-like phase.

"Fundamental work is underway to understand the exact nature of this Kondo-like phase, whether it has any topological properties, and whether it connects to structural defects," Yang said. "Eventually we want to understand how to microscopically describe such an exotic system."

For portfolio planners, the near-term signal is a stoichiometric route to interaction-driven flat bands that bypasses Moiré-stack manufacturing. Groups already evaluating van der Waals magnets for spintronic or memory research should track whether the same coherent state survives above cryogenic conditions, and whether the laser-switching scheme can deliver reproducible bit writing once that barrier falls.