Frozen Chiral Phonons Give Antiferromagnets a Giant Spin-Split Surprise

Antiferromagnets have long been the quiet workhorses of magnetism: their atomic spins cancel in perfect opposition, making them invisible to stray magnetic fields and therefore attractive for dense, fast, low-power electronics. But that same cancellation has been a curse for engineers, because a material with no net magnetization normally refuses to produce the spin-dependent phenomena that make ferromagnets useful. A team of researchers led by Su-Yang Xu and Philip Kim at Harvard University, together with collaborators across more than twenty institutions, now reports in Nature that a seemingly ordinary collinear antiferromagnet can be coaxed into producing some of the largest topological responses ever measured in a bulk magnet, simply because it secretly harbors a hidden chiral superlattice in its crystal structure.

The material in question is uranium oxytelluride, or UOTe, a layered van der Waals compound that orders antiferromagnetically below a Néel temperature of roughly 150 kelvin. In its idealized crystallographic description, UOTe is about as boring as a magnet can be: the spins line up in straight, antiparallel rows, and the electronic bands are spin-degenerate with vanishing Berry curvature, the quantum-geometric quantity that underlies exotic transport effects such as the anomalous Hall effect. The Harvard-led team, however, discovered that the real crystal refuses to obey this tidy textbook picture. Using transmission electron microscopy, they found that the uranium atoms spontaneously twist into a slowly varying, spiral-like displacement pattern, forming a superlattice with a periodicity several times larger than the nominal unit cell.

The origin of this supermodulation is remarkable in itself. First-principles calculations show that a branch of the phonon spectrum, the quantum vibrations of the crystal lattice, goes soft at a finite wave vector, meaning the lattice lowers its energy by freezing that vibration into a static distortion. The frozen mode is chiral: the uranium atoms wind in a handed, screw-like fashion, with two sublattices winding in opposite senses. The result is a two-dimensional planar chirality woven into the atomic positions, a structural motif previously engineered only artificially in metamaterials and molecular assemblies. In UOTe, nature builds it spontaneously, and the electron diffraction patterns reveal a monoclinic shear angle of about 1.23 degrees that accumulates layer by layer, a direct fingerprint of the collective twist.

Why does this matter for electrons? When an electron’s Bloch wavefunction travels through the crystal, it experiences the chiral superlattice potential as a slowly varying modulation of its orbital character and hopping amplitudes. The team’s tight-binding modeling shows that this modulation warps the quantum geometry of the electronic bands, generating a large Berry curvature and a substantial Berry curvature dipole even though the underlying magnetic order remains perfectly collinear. In other words, the geometry of the lattice, not the arrangement of spins, becomes the source of topology. This is a conceptual inversion of the usual recipe, where strong spin-orbit coupling or non-collinear spin textures are required to produce Berry curvature in a magnet.

The experimental smoking gun came from the nonlinear Hall effect, a second-order electrical response that appears under time-reversal-symmetric conditions and directly probes the Berry curvature dipole. The researchers observed a clear nonlinear Hall voltage in UOTe devices, confirming that the chiral superlattice endows the bands with the predicted geometric structure. But the most dramatic result emerged when this Berry curvature was allowed to couple to the antiferromagnetic order itself. Below the Néel temperature, the material develops an anomalous Hall effect with a Hall angle of approximately 0.14, meaning that roughly one in seven of the transverse charge carriers is deflected by the topological response, a figure among the largest ever recorded in any bulk magnet, ferromagnetic or antiferromagnetic.

Careful control experiments ruled out mundane explanations. Nitrogen-vacancy center magnetometry and magnetic circular dichroism measurements showed that any net magnetization in the sample is tiny, on the order of ten milli-Bohr magnetons per uranium ion, far too small to account for the Hall signal. The team demonstrated that the large anomalous Hall response persists at zero magnetization and that the intrinsic antiferromagnetic contribution dominates over both the ordinary Hall effect and any weak ferromagnetic component. The abrupt onset of the effect near the Néel temperature further ties the response to the magnetic order, establishing that the chiral superlattice and the collinear antiferromagnetism act synergistically rather than independently.

Perhaps the most consequential finding for technologists is the generation of a spin-polarized current from the collinear antiferromagnet. Using spin Hanle precession measurements, a technique borrowed from graphene and semiconductor spintronics in which spin precession in an external field reveals the spin content of a charge current, the team showed that electrons flowing through UOTe carry a measurable spin polarization. Extracting spin currents from antiferromagnets has been a long-standing goal of the field, because antiferromagnetic spintronic devices would combine terahertz-scale switching dynamics, immunity to magnetic perturbations, and compatibility with existing fabrication methods. The chiral superlattice provides a symmetry-breaking mechanism that makes this possible without relying on heavy-element spin-orbit torques or adjacent ferromagnetic layers.

Beyond the single material, the work delivers a design principle rooted in chemistry. The team explains the supermodulation as a compromise between competing energy scales: in the isostructural compound UOS, an additional interlayer bond stabilizes a three-dimensionally bonded structure and no superlattice forms, whereas in UOTe the long uranium-tellurium bonds make such bonding costly, so the layers locally curve up and down, opening bond angles at some locations to enable interlayer bonding while compressing them elsewhere, producing intralayer repulsion. The balance between these interlayer and intralayer energies yields the frozen chiral phonon. By parameterizing this balance through ionic radii, bond angles, and bond-length ratios, the researchers identified empirical rules that can be applied to the roughly five hundred compounds in the Inorganic Crystal Structure Database that are isostructural to UOTe, pointing the way to a family of bond-mismatch chiral superlattices.

The conceptual implications extend into the rapidly growing field of altermagnetism and topological antiferromagnetism, where researchers seek magnetic materials whose spin-split bands resemble those of ferromagnets without carrying net magnetization. UOTe demonstrates a route that bypasses the usual symmetry requirements entirely: instead of engineering spin splitting through the magnetic space group alone, one can sculpt the quantum geometry of electrons through real-space structural modulation. Because the superlattice strongly modifies local bonding environments rather than merely straining the lattice, it provides a direct handle on orbital composition and hopping amplitudes, the ingredients from which Berry curvature is built. This bond-mismatch strategy complements lattice-mismatch approaches such as moiré superlattices, and it operates in a chemically stable bulk crystal rather than a delicate twisted bilayer.

For the broader quantum materials community, the study suggests that hidden chiral supermodulations may be more common than appreciated, lurking in layered compounds whose average structures look deceptively simple. The combination of spontaneous chirality, collinear antiferromagnetism, giant anomalous Hall response, and spin-polarized transport in a single van der Waals crystal offers a unusually rich platform for exploring how real-space geometry programs momentum-space topology. If the design principles hold across the predicted chemical space, the coming years could see a deliberate search for chiral superlattice antiferromagnets that operate at and above room temperature, bringing topological antiferromagnetic spintronics a decisive step closer to practical devices.

Subject of Research: Chiral superlattice formation in the collinear antiferromagnet UOTe and its generation of spin-split topological transport

Article Title: A chiral superlattice route to spin-split topological antiferromagnetism

Article References: Dinh, T., Liu, M., Qiu, J.-X., Mardanya, S., Sung, S. H., Le, X. H., Broyles, C., Zhu, C., Xu, Q., Chen, H., Rehfuss, Z., Liu, Y.-F., Zeng, X., Li, H., Guo, P., Liu, J., Huang, T., Shi, J., Smith, M., … Xu, S.-Y. (2026). A chiral superlattice route to spin-split topological antiferromagnetism. Nature, 658(8135), 342-349. https://doi.org/10.1038/s41586-026-11073-7

Image Credits: AI Generated

DOI: 10.1038/s41586-026-11073-7

Keywords: antiferromagnetism, chiral superlattice, UOTe, Berry curvature, anomalous Hall effect, frozen chiral phonons, spintronics, nonlinear Hall effect, topological materials, spin-polarized current, quantum geometry, van der Waals materials