For nearly two decades, the condensed matter physics community has searched for definitive experimental proof that a quantum spin liquid (QSL) can exist in a real material. A paper published June 10 in Nature Physics by an international team led by the University of Oxford, the University of Bristol, and University College Cork may have delivered it, by turning a long-standing problem in the field into its greatest advantage.
Herbertsmithite (ZnCu₃(OH)₆Cl₂), a mineral first fully synthesized in the laboratory in 2005, has been the leading candidate for hosting a QSL ground state. Its crystal structure features perfect kagome planes, a two-dimensional lattice of corner-sharing triangles, of spin-½ copper ions (Cu²⁺), separated by non-magnetic zinc (Zn²⁺) planes. The geometry is inherently frustrated: no configuration of the copper spins can simultaneously satisfy all the magnetic interactions, and the system is forced into a highly entangled, non-magnetic ground state where the electron spins never order, even at temperatures approaching absolute zero.
The theoretical signature of such a state is the existence of spinons, fractionalized quasiparticles that carry spin-½ but no electric charge. Unlike the spin waves (magnons) that propagate through ordinary magnets, spinons are the elementary excitations of a QSL. Detecting them directly has been the central challenge.
The impurity problem
Real herbertsmithite crystals are not perfect. Approximately 5-10% of the zinc sites are inevitably occupied by copper atoms, a form of chemical disorder that has been viewed as a nuisance for the entire history of the field. These “defect” copper spins, sitting in the non-magnetic zinc layers, interact with the kagome planes and produce low-temperature magnetic behavior (a Curie-Weiss tail in susceptibility, a specific heat anomaly) that has been difficult to reconcile with any QSL model.
Many skeptics argued that what appeared to be QSL behavior in herbertsmithite was actually just the collective physics of these impurity spins freezing into a conventional spin glass.
The team, led by co-first authors Hiroto Takahashi (Oxford), Jack Murphy (UCC Cork), and Mitikorn Wood-Thanan (Cardiff/Bristol), and corresponding authors Felix Flicker (Bristol) and J. C. Séamus Davis (Oxford/UCC/MPI Dresden/Cornell), took a different view. Rather than treating the impurity spins as contamination, they reconceptualized them as quantum witness spins, local probes whose fluctuating dynamics reveal the presence and properties of the spinons in the kagome planes, much as a witness to a crime can testify about events they observed.
Spin noise spectroscopy
To read the testimony of these witness spins, the team turned to a relatively uncommon technique: spin noise spectroscopy (SNS) . Unlike neutron scattering or nuclear magnetic resonance, the conventional workhorses of quantum magnetism research, SNS measures spontaneous fluctuations in magnetization from a sample at thermal equilibrium, without perturbing it. A laser beam is passed through the crystal, and the rotation of its polarization due to the fluctuating magnetization of the witness copper atoms is recorded over time.
The team performed these measurements at millikelvin temperatures using a custom-built spectrometer. What they found was unexpected.
A sharp cusp appeared in the DC magnetic susceptibility at T* ≈ 260 mK, a phase transition. Below this temperature, the witness spins entered a spin glass phase, confirmed by characteristic aging effects: the dynamics depended on the sample’s thermal history, a hallmark of glassy systems.
This, by itself, would seem to confirm the skeptics’ view. But the team went further. They developed a theoretical model in which the witness spins interact not directly with each other, but through the spinons propagating in the kagome planes. The spinons mediate an effective interaction between witness spins, an interaction that becomes stronger as temperature decreases, eventually driving the spin glass transition.
“It’s like two buoys on the ocean connected by waves passing between them,” explains Flicker. “The waves are the spinons, and the buoys are the witness spins. By watching the buoys move, you can infer the existence and properties of the waves.”
Evidence for a Z₂ QSL
The model fit the data remarkably well. With only one free parameter, it simultaneously explained the spin glass transition temperature, the temperature dependence of the susceptibility, the frequency and temperature structure of the noise spectrum, the Curie-Weiss temperature (~-4 K), and even previously published neutron scattering data that had been difficult to interpret.
Critically, the team tested both categories of QSL that theory permits. A U(1) QSL (with gapless spinon excitations) and a Z₂ QSL (with a small energy gap) both produce spinon-mediated interactions between witness spins. But the Z₂ model provided the significantly better fit, suggesting that herbertsmithite’s ground state is a Z₂ quantum spin liquid, one of the most theoretically pristine forms of topological order in a magnetic system.
What this means
The witness-spin approach is more than just a result about one material. It provides a general methodology for studying QSL candidates that have been dismissed as “too dirty” to reveal their quantum secrets. The same technique could be applied to other frustrated magnets, potentially revealing a broader family of fractionalized materials than previously recognized.
The paper, which is open access, includes 15 authors from 10 institutions across the UK, Germany, and the United States, including the Max Planck Institute for Solid State Research in Stuttgart and Argonne National Laboratory.
The result does not close the book on quantum spin liquids. The structural connectome of the kagome planes themselves, the direct observation of spinon dynamics within them, remains an open challenge that may require new experimental techniques. But the witness-spin method has provided what many in the field consider the most compelling evidence to date that herbertsmithite is indeed the long-sought quantum spin liquid.
Source: Takahashi, H., Murphy, J., Wood-Thanan, M. et al. “Spinon mediation of witness spin dynamics in herbertsmithite.” Nature Physics (2026). DOI: 10.1038/s41567-026-03303-6. Published 10 June 2026. Open access. arXiv: 2510.11678.