- Home
- News
- General News
- RIXS shows fingerprint...
RIXS shows fingerprint of Kitaev magnetism
08-10-2026
Experiments at the ESRF’s ID20 beamline have revealed a distinctive form of magnetic excitation in a material long studied for its unusual quantum magnetism. Using RIXS interferometry, researchers have directly observed spin excitations that differ according to bond direction, even between directions that are equivalent in the crystal structure – a defining feature of so-called Kitaev magnetism. The results are published in npj Quantum Materials.
Share
In conventional magnets, the spins of electrons interact with one another to form an ordered state, such as the parallel alignment found in a ferromagnet. Physicists have long sought more exotic “quantum spin liquids”, in which strongly interacting spins remain fluctuating and entangled, even at temperatures close to absolute zero.
One possible route to such a state was proposed by the physicist Alexei Kitaev. In the Kitaev model, each spin is connected to three neighbours – as occurs, for example, when spins sit on the vertices of a honeycomb lattice. Crucially, the way any two spins interact depends on the direction of the bond connecting them. Such “bond-directional” interactions can frustrate conventional magnetic order and, in the ideal Kitaev model, produce a quantum spin liquid. The ideal Kitaev spin liquid hosts unusual collective excitations known as Majorana fermions, which have attracted interest for their potential applications in quantum computing and other quantum technologies.
Finding this physics in real materials has proved difficult. Na₂IrO₃ has attracted particular interest because strong spin–orbit coupling is predicted to produce bond-directional interactions. It is not itself a spin liquid – at low temperatures, its spins form an ordered zigzag pattern. But it has been proposed as a “proximate” spin liquid, in which Kitaev physics can persist amid other magnetic interactions.
Now Markus Grüninger at the University of Cologne in Germany and colleagues have directly detected Kitaev-type, bond-directional excitations in Na₂IrO₃. “They do exist – and we can measure them with RIXS!” says Grüninger.
The team used an approach known as RIXS interferometry, which Grüninger and colleagues have developed over several years in experiments at ID20. Rather than measuring RIXS spectra at only a few points in momentum space, the technique follows changes in X-ray scattering intensity over a large range of momentum transfer. By analysing interference patterns in this scattering intensity, physicists can identify correlations between pairs of neighbouring atoms.
In this latest work, Grüninger and colleagues combined this approach with a sophisticated use of X-ray polarisation, tilting a sample of Na₂IrO₃ by precisely 54.7° while taking measurements at a scattering angle of 90°. The carefully chosen geometry allowed the researchers to distinguish two different kinds of magnetic excitation – and reveal differences between excitations along bonds that are otherwise equivalent in the crystal structure, exposing their bond-directional character.
According to the researchers, the existence of bond-directional excitations provides evidence of bond-directional magnetic interactions in Na₂IrO₃ – although they stress that the pure Kitaev model alone cannot account for both excitations, and the experiment does not yet quantitatively determine the Kitaev interaction strengths. But, says Grüninger, “the existence of bond-directional excitations tells us that exchange interactions have a strong bond-directional character.”
Because RIXS produces relatively few scattered photons, data collection is lengthy, and experiments must be set up with very high stability. According to Grüninger, the observation of the bond-directional excitations has pushed the technique to new limits. “What we do is at the edge of what is possible,” he says.
The results could provide a new way to investigate candidate Kitaev materials. In the past, physicists have sometimes had to infer strong bond-directional interactions from the particular form of magnetic order that a material adopts. That approach cannot be applied to a true quantum spin liquid, because such a state does not develop long-range magnetic order in the first place. By probing bond-directional excitations instead, RIXS interferometry offers another route to identify Kitaev interactions in candidate materials.
The researchers now want to develop their RIXS approach further, with the goal of extracting quantitative information about the exchange interactions in candidate Kitaev materials. “We are working on an even more sophisticated version of the experiment to make progress in that direction,” says Grüninger.
Reference:
Text by Jon Cartwright
Top image: The setup on ID20. Credits: Ava Rajh (ID20 post-doc).



