Frustratingly disordered

A study of how disorder affects a ‘frustrated’ magnet reveals a surprising robustness of the underlying quantum many-body state, and provides evidence for emerging quantum phenomena induced by disorder.

In a frustrated magnet, spins are arranged in geometries that prevent the formation of conventional magnetic patterns such as ferromagnetic or anti-ferromagnetic order. Yet, the interaction between the spins can nonetheless lead to highly correlated quantum states, including exotic ones that remain not fully understood such as so-called spin ices or spin liquids. Frustrated magnets are therefore ideal materials for investigating, both theoretically and experimentally, the intricate interplay of interactions in quantum many-body systems and the phases emerging from them. At the forefront of such explorations are rare-earth pyrochlores, which are frustrated magnets that have been extensively studied over the past decade or so. But what happens when these materials are affected by disorder in the crystal structure is largely unexplored in experiments, despite theoretical predictions of intriguing phenomena emerging from disorder. This is the gap Romain Sibille and colleagues at the Laboratory for Scientific Developments and Novel Materials and the Laboratory for Neutron Scattering and Imaging started to fill now with a study that appeared today in Nature Communications [1].

Surprisingly robust

Basic geometry of the corner-sharing tetrahedra making up the network of magnetic ions (here: terbium, Tb) in pyrochlore materials.
In pyrochlores, magnetic ions form a network of corner-sharing tetrahedra, a configuration that prevents the system from reaching an ordered ground state. As the frustration depends on the geometry of the crystal lattice, one might expect that any structural change has an immediate effect on the properties of the material. All the more it came as a surprise when Sibille and his co-workers now found that even very strong disorder does little to change the magnetic behaviour of the system — at least for the sort of disorder they studied.

The PSI scientists worked with the pyrochlore magnet Tb2Hf2O7. For this study they produced a single-crystal sample of this material in which around 8% of the oxygen atoms (O) are relocated in the structure. This form of disorder leaves the network of terbium (Tb) tetrahedra — which are responsible for the magnetic properties — fully intact, but for around half of them one O atom is missing. As some of the bonds between the Tb ions involve oxygen (Tb–O–Tb), the overall bond network is therefore directly affected. Moreover, the symmetry of the environment around the Tb sites is broken, which should in principle make the ions non-magnetic. This is, however, not what Sibille et al. observed in their experiments involving X-ray and neutron diffraction, muon spin relaxation and other characterisation techniques.

New phenomena emerging

Spin-flip (left) and non-spin-flip (right) scattering maps of the disordered Tb2Hf2O7 crystal measured using neutron polarization analysis. (From [1].)
Instead of a loss of magnetic properties they found that the ions remain magnetic and correlated, the high degree of disorder notwithstanding. That the material maintains its magnetic properties hints towards a stabilisation of the magnetic phase through the structural defects. Moreover, the PSI researchers observed the emergence of a new phenomenon, a spin-glass transition, which might be due to the bond disorder introduced in the system. These findings are consistent with recent theoretical predictions that disorder in pyrochlore systems might give rise to qualitatively new quantum phenomena. And given that in the approach now introduced by Sibille and colleagues the level of disorder can be controlled during sample preparation, their material holds the promise to serve as a flexible platform for exploring more broadly the intriguing physics of disordered frustrated magnets.

This work was carried out in a collaboration led by Romain Sibille and Michel Kenzelmann (PSI), with partners at the Swiss Light Source and the Laboratory for Muon Spin Spectroscopy (PSI), the University of Warwick (UK), the Institut Néel and Institut Laue-Langevin in Grenoble (France), the Rutherford Appleton Laboratory in Didcot (UK) and the Oak Ridge National Laboratory (US).