Cool a magnetic material down far enough and its ions have to make a choice. Ferromagnetic, lining up in the same direction like soldiers called to attention, or antiferromagnetic, alternating north-south-north-south like a perfectly argued disagreement. At temperatures within a whisker of absolute zero, most materials settle into one or the other, the system picking a side and locking in. What the field has been watching for, hunting through material after material, is something different: a substance that refuses to choose at all, that holds both possibilities in suspension simultaneously through the strange arithmetic of quantum mechanics. That would be a quantum spin liquid.
The hunt matters because quantum spin liquids are, theoretically, remarkable things. The physics that keeps them perpetually undecided involves fractionalized excitations, quasi-particles that behave as though each electron’s spin has split apart from the rest of its properties and gone wandering independently through the material. That fractionalization is potentially useful for quantum computation in a particularly attractive way: it’s fault-tolerant, meaning errors don’t cascade. Finding one would be a genuine coup.
So when cerium magnesium hexaaluminate (CeMgAl11O19) turned up in the literature displaying the two main signatures, researchers reasonably called it a quantum spin liquid candidate. It showed a continuum of different states rather than a single settled configuration. It showed no magnetic ordering. Both boxes ticked. A team at Rice University, led by Pengcheng Dai, decided to look closer. What they found was something the field had not described before.
The material was not a quantum spin liquid. Not even close, it turned out.
“The material had been classified as a quantum spin liquid due to two properties: observation of a continuum of states and lack of magnetic ordering,” said Bin Gao, a research scientist at Rice and co-first author on the paper, published this week in Science Advances. “But closer observation of the material showed that the underlying cause of these observations wasn’t a quantum spin liquid phase.”
The difference lies in what’s happening at the level of individual ions. In CeMgAl11O19, the boundary between ferromagnetic and antiferromagnetic states happens to be weaker than in most comparable materials, meaning the cerium ions have more flexibility to switch between the two orientations. The result is a structure where some ions are ferromagnetic and some are antiferromagnetic at the same time, not because quantum mechanics demands it but because the competing interactions are so evenly balanced that neither wins. This produces the lack of magnetic ordering. And because no single ordered state is selected, the system has an unusually broad landscape of available low-energy configurations, which, when probed, looks very much like the continuum signature that would indicate fractionalized quantum excitations.
In a true quantum spin liquid, the system can shift between those low-energy states via quantum tunnelling, flickering between configurations through genuinely quantum processes. In CeMgAl11O19, it can’t. Once it settles into a configuration, it stays there. The continuum that appears in measurements is not a quantum superposition but an ensemble average, the aggregate of a population of slightly different ordered states all present simultaneously across the material’s bulk. Thermally disordered, not quantum. “It was not a quantum spin liquid, yet we were observing what we thought were quantum spin liquid-associated behaviors,” said co-first author Tong Chen.
The team established this through neutron scattering at multiple facilities, including the Spallation Neutron Source at Oak Ridge and the J-PARC accelerator complex in Japan, bombarding the material and mapping how the neutrons scattered to reconstruct the underlying spin dynamics. The crucial step was measuring the material in a strong magnetic field (4 tesla along the crystal axis, enough to force all the ions into alignment) and then analysing the sharp, well-resolved spin waves that emerged. Those field-polarised measurements allowed Dai’s team to precisely determine the material’s spin Hamiltonian, which in turn placed CeMgAl11O19 close to an exactly solvable point in a theoretical phase diagram where ground-state degeneracy is expected to be extensive. The calculated ensemble average of all those degenerate states, when fed into modelling software, reproduced the observed neutron spectrum quantitatively. A classical explanation, with no quantum fractionalization required.
This matters well beyond one compound. Quantum spin liquids remain the most exotic state in condensed matter physics, and identifying genuine examples is extraordinarily difficult. The two hallmarks of the search, the continuum and the absence of magnetic ordering, are not specific enough. Other materials, notably YbMgGaO4, have faced similar scrutiny over whether their signatures are genuinely quantum or the product of structural disorder producing similar-looking spectra through mundane means. The Rice result provides a specific, well-characterised counter-example: a mechanism that can produce QSL-like signals without any quantum spin liquid physics. “The material’s unique ability to ‘choose’ between different low energy states produced observational data very similar to a quantum spin liquid state,” Dai said.
The claim is a substantial one, and arguably the more arresting part of what the team is proposing. Not just that this material was misidentified, but that it represents something without a prior name. “This is a new state of matter that, to our knowledge, we are the first to describe,” said Dai. A mixed ferro-antiferromagnetic degenerate state, stabilised by a combination of geometric frustration, competing exchange interactions, and just enough disorder to prevent the system from picking a single configuration even as temperatures approach absolute zero.
That last point, the role of disorder, is perhaps the subtlest. In an ideal system near the degenerate point, quantum fluctuations might eventually lift the degeneracy through a mechanism called order-by-disorder, nudging the material into a unique ground state at the very lowest temperatures. The roughly 7% vacancy in CeMgAl11O19’s cerium sites, a defect present in similar hexaaluminate compounds, seems to prevent this. The disorder doesn’t destroy the degenerate landscape; it preserves it, freezing the system into its inconclusive standoff. Which means that what looked like evidence of the exotic has turned out to be evidence of something more mundane and, in its way, rather stranger: a material permanently suspended between orderings by its own imperfections.
Dai is circumspect about the broader implications in the measured way of someone who has just spent a long time demonstrating what a material is not. “It underscores the importance of careful observation and thorough investigation of your data.” The work will not simplify the quantum spin liquid search; if anything, it adds an item to the checklist of alternative explanations to rule out before making the QSL claim. But it does describe the landscape more honestly. And the new state itself, the one the team stumbled into while looking for something else, may turn out to be stranger territory than the quantum physics it was initially mistaken for.
DOI / Source: https://doi.org/10.1126/sciadv.aed7778
Frequently Asked Questions
If this material isn’t a quantum spin liquid, does that mean quantum spin liquids don’t exist? Not at all. Quantum spin liquids remain a well-supported theoretical category and several candidate materials are still actively studied. What this research demonstrates is that two of the most widely used experimental signatures for identifying them, a continuum of spin excitations and the absence of magnetic ordering, are not on their own sufficient proof. The criteria need to be applied more carefully, and this work provides a specific alternative mechanism that can produce the same signals without the underlying quantum physics.
What makes a quantum spin liquid useful for quantum computing, and does this finding affect that? The appeal of quantum spin liquids for fault-tolerant quantum computation lies in their fractionalized excitations: quasi-particles whose properties are distributed across the material in ways that make local disturbances less likely to corrupt stored information. CeMgAl11O19 doesn’t have those excitations, so it doesn’t carry that potential. The finding doesn’t set back quantum computing directly, but it does mean the field needs sharper tools for confirming which materials genuinely have the physics required.
How did neutron scattering expose what other measurements missed? Neutron scattering can map the energy and momentum of magnetic excitations across a material with high resolution, revealing the internal spin dynamics rather than just their bulk magnetic properties. The key advance here was measuring the material both in zero field and in a strong applied magnetic field that forced all the ions to align. In the polarised state, sharp well-defined spin waves emerged, and their shape allowed the team to pin down the precise exchange interactions governing the material. That Hamiltonian, once known, predicted the zero-field spectrum exactly using a classical ensemble average, closing the quantum interpretation off.
Why would disorder in a crystal actually stabilise an unusual state rather than destroying it? Usually disorder is a nuisance, disrupting ordered phases and producing noise in measurements. Here it plays an oddly constructive role. The material sits near a point in its phase diagram where many configurations have almost identical energy. In a perfect crystal, even tiny quantum fluctuations would eventually select one of those configurations and produce long-range order. The cerium vacancies and ion-mixing present in this material prevent that selection from happening, locking the system into a coexistence of multiple states that persists all the way to the lowest temperatures measured, around 100 millikelvin. Disorder, in this case, is what keeps the unusual state alive.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.