SLAC Researchers Confirm Altermagnetism as a Third Magnetic Phase

For nearly 90 years, physics textbooks have divided magnetic materials into exactly two categories based on how their internal electrons behave. On October 29, 2026, a research team led by the SLAC National Accelerator Laboratory shattered that binary. Writing in the journal Science, the scientists delivered conclusive experimental proof of a third magnetic phase.[1][5]

The newly confirmed phase is called altermagnetism. By firing highly controlled X-rays at a sample of manganese telluride, the researchers detected a distinct structural twist that conventional theories cannot explain. The finding validates a theoretical model first proposed in 2022.[1][2]

"In the past few years, scientists have observed phenomena that don't fit neatly into the ferromagnetism-antiferromagnetism dichotomy," explained Daniel Jost, the study's lead author. Jost serves as an associate scientist at SLAC and the Stanford Institute for Materials and Energy Sciences.[1]

"Researchers have been compiling evidence of a third type of magnetism that bridges the qualities of these two classes," Jost added. "We believe our results provide robust experimental confirmation that this third phase—altermagnetism—is real."[1]

The Century-Old Binary

To understand why a third phase matters, engineers look to the two established forms of magnetism that occur without external influence. The first is ferromagnetism, the force behind refrigerator magnets and electric motors. In a ferromagnet, every electron spin points in the exact same direction, creating a strong, observable magnetic field.[2][3]

That unified alignment makes ferromagnets easy to read, which is why they form the basis of modern hard drives. However, their external magnetic field makes them highly sensitive to outside interference. A stray magnetic field can easily flip the spins, scrambling the stored data.[2]

The second category, antiferromagnetism, was discovered in 1933. In these materials, the electron spins alternate perfectly—one points up, the next points down, and so on. Because the opposing spins cancel each other out, an antiferromagnet produces zero net magnetic field.[2]

This cancellation makes antiferromagnets incredibly stable and immune to outside magnetic interference. But that same stability makes them practically invisible to sensors. Engineers cannot easily read the data stored within them or use them to drive the efficient spin currents required for next-generation computing.[2]

The Spintronics Bottleneck

This binary classification created a stubborn bottleneck for a field called spintronics. Traditional electronics process information using the electrical charge of electrons. Spintronics attempts to use both the charge and the internal spin of the electron to store and move data, promising faster and vastly more energy-efficient devices.[3]

"If you have a permanent magnet in your pocket and walk by a spintronic device built with ferromagnets, you could start flipping the bits, scrambling the information stored there," Jost noted. Antiferromagnets fix the stability problem, but their invisible nature makes them nearly useless for extracting data.[1]

Engineers needed a material that combined the best traits of both: the zero net magnetization of an antiferromagnet for stability, and the readable spin-polarized currents of a ferromagnet. In 2022, theorists proposed that such a material might already exist, hiding in plain sight. They called it an altermagnet.[1][2]

A Geometric Twist

The 2022 theoretical framework suggested that altermagnets share the alternating up-down electron spins of an antiferromagnet, resulting in zero net magnetic field. However, the theory proposed a crucial geometric twist hidden within the material's atomic architecture.[1]

In an altermagnet, the actual crystal lattice surrounding the "up" spins is rotated relative to the lattice surrounding the "down" spins. This structural rotation breaks the material's symmetry, giving its electronic and magnetic structure a distinct handedness, or chirality.[1][3]

Human hands are a classic example of chirality: a left hand and a right hand are mirror images, but a left hand cannot fit properly into a right-handed glove. If researchers could prove that a magnetic material possessed this chiral twist, they would prove that altermagnetism was real.[1]

"By 2024, there were some factors that spoke strongly in favor of altermagnetism being real," Jost said. "But there was still a missing piece to the puzzle." That missing piece was direct, physical observation of chirality in a candidate material.[1]

Hunting for Handedness

To find that proof, the SLAC-led team focused on manganese telluride. For decades, the scientific community had classified the compound as a standard, run-of-the-mill antiferromagnet. Recent theoretical models, however, flagged it as a prime candidate for altermagnetism.[1][3]

The researchers took their samples to the European Synchrotron Radiation Facility in France. They utilized the facility's ID32 beamline, an instrument capable of generating highly controlled, circularly polarized X-rays. This specific technique is known as circular-dichroism resonant inelastic X-ray scattering.[1]

AI

The testing method relied on a simple principle: chiral materials scatter left-handed polarized X-rays differently than they scatter right-handed beams. A symmetrical, non-chiral material—like a conventional antiferromagnet—would scatter both beams in the exact same pattern.[1]

When the team fired the polarized beams at the manganese telluride, the material scattered the left and right twists into different patterns. The material's internal geometry could distinguish between the two directions of light, revealing a clear chiral signal.[1][3]

Confirming the Third Phase

"This chiral signature gave us the solid experimental ground to say altermagnetism stands in a class of its own," Jost said. The distinct handedness proved that the crystal lattice was indeed rotated, confirming the 2022 theoretical models.[1]

The team completed their initial experiment in 2024. To ensure the accuracy of their groundbreaking result, they returned to the French synchrotron facility in early 2025 and repeated the exact same measurements. The chiral signature appeared again, perfectly reproducing the initial data.[1]

"This technique helped us confirm that manganese telluride was really exhibiting altermagnetic properties," said Rachel Nickel, a postdoctoral scientist with ESRF who helped design the ID32 beamline experiment. "Altermagnetism was only proposed a few years ago, and the rapid progress in confirming these models is incredible."[1]

The results were formally published in Science under the title "Chiral altermagnon in MnTe." The paper lists 15 co-authors from international institutions, including Stanford University, the Technical University of Munich, and the Walther-Meissner-Institute.[5]

The Future of Memory

The confirmation of altermagnetism provides engineers with a new degree of freedom. Because altermagnets possess zero net magnetization, they remain perfectly stable against outside magnetic interference. Yet, because of their internal twist, they can still generate the spin-polarized currents necessary to extract data.[1][2]

This combination makes them ideal candidates for next-generation memory devices, quantum technologies, and highly efficient data centers. By utilizing electron spin rather than just electrical charge, future computers could operate at vastly higher speeds while consuming a fraction of the power.[1][3]

However, the researchers caution that moving from a laboratory proof of concept to a commercial semiconductor facility will require years of additional engineering. Future research will focus on how these altermagnetic states can be manipulated and controlled using light and electricity.[1]

For now, the discovery represents a rare, fundamental shift in materials science. A classification system that stood unchallenged for nearly a century has been rewritten, opening an entirely new branch of physics just four years after it was first theorized.[1][2]