Chameleon Altermagnet: Two Games on One Chessboard ✛ ✦

‹›

In the world of magnetism, a simple dichotomy long reigned: ferromagnets, where spins point in one direction, and antiferromagnets, where they line up in a checkerboard pattern, canceling each other out. But a recent 'third player' — altermagnets — forced a rewrite of the rules. In these materials, spins still form an antiferromagnetic pattern, but due to specific crystal lattice symmetry (here Noether's theorem, discovered by Emmy Noether, comes into play), electrons experience something strange: bands with different spin polarizations diverge in momentum space, like ripples from two stones thrown at once. KV₂Se₂O became a promising altermagnet candidate, but its magnetic structure turned out to be a master of disguise — momentum-based methods such as photoemission spectroscopy gave conflicting signals.

To resolve the dilemma, physicists turned to direct vision. Imagine a layered crystal as a deck of playing cards. Each card — an atomic layer of vanadium oxide — is painted with a checkerboard spin pattern: black square — spin up, white — spin down. But how is the deck stacked? Cards can lie on top of each other repeating the same pattern, or alternate as mirror reflections, like a trick with flipped cards. The first option (C-type) yields an altermagnetic phase, the second (G-type) — an ordinary antiferromagnet. The problem is that the 'paperclips' connecting the layers — the interlayer exchange interaction — are negligibly weak: about 1 meV. That’s twenty times less than the thermal energy at room temperature, so the deck becomes pliable: the slightest breeze — a local strain — and the magnetic order flips.

Researchers from the Institute of Physics of the Chinese Academy of Sciences used spin-polarized scanning tunneling microscopy (the idea of tunneling was once explained by John Bardeen, and Leo Esaki discovered the tunnel effect). Cooling the sample to ultra-low temperatures where decoherence is suppressed, they passed a current through a sharp spin-polarized probe and scanned the surface. The method of quasiparticle interference (QPI) allowed them to reconstruct the electronic structure from the pattern of standing waves scattered off defects — just as ripples on water betray a submerged rock. By changing the probe's magnetization, the physicists read out the spin texture of each area, like flipping page after page — and each page turned out to be a magnetic layer.

The result was stunning. On different terraces of the same sample — most often on cleavage steps, those accidental peepholes into the next layer — sometimes identical, sometimes opposite spin patterns were observed. This meant direct coexistence of two phases on the surface. First-principles calculations (DFT) confirmed: the energies of the C-type and G-type configurations are separated by only about 1 meV per atom, so the slightest local strains or defects tip the scales one way or the other. Hence the contradictions of past years: momentum spectroscopy saw an averaged picture, while real space held a dualism — like a two-faced Janus, the crystal showed one face or the other, depending on where you looked.

The discovery forces a fresh look at the hunt for altermagnets. 'Bird's-eye views' — photoemission measurements — alone are not enough. It is necessary to descend to the atomic level and measure the spin order directly. This is especially crucial for future spintronic devices, where magnetic symmetry dictates functionality. If we can learn to switch configurations with a light nudge — voltage, pressure, or light — then altermagnets will become the basis for non-volatile memory and logic elements based on pure spin currents. Nature loves ambiguity on the verge of indistinguishability, and it is in these microscopic gaps that tomorrow's technologies are born.