Textbooks have long split permanent magnetism into two families. A team led by SLAC National Accelerator Laboratory now says the list needs a third entry. In a paper published in Science on 29 October, the researchers report what they call conclusive experimental proof of altermagnetism, a phase that theorists only named in 2022.
The difference comes down to how electron spins line up. In a ferromagnet, the kind that pins drawings to a fridge, they all point the same way. Antiferromagnets flip the direction from one atom to the next, so the effects neutralise each other. Altermagnets share that up-down rhythm, yet the crystal surroundings of each orientation are turned at an angle to one another. Theory says this geometric quirk should leave a mirror-image asymmetry, known as chirality, which a standard antiferromagnet does not have.
Lead author Daniel Jost and colleagues went looking for that handedness in manganese telluride, a compound treated as a plain antiferromagnet for decades. At the European Synchrotron Radiation Facility in France they fired circularly polarised X-rays at the crystal, alternating between left- and right-handed beams. The material scattered the two differently, a chiral fingerprint that a conventional antiferromagnet should not show. The first run in 2024 was repeated in early 2025 with the same outcome.
For Jost, the signal shows that altermagnetism “stands in a class of its own”. The practical interest lies in spintronics, which stores and moves information using electron spin as well as charge. Ferromagnets are easily scrambled by stray fields, while antiferromagnets are hard to read out. Altermagnets promise both stability and usable spin currents.
The team now wants to learn how to switch these states with light and electric currents, with an eye on memory chips, quantum technologies and more energy-efficient data centres.
