Defects Drive Disorder
The diffuse scattering measured from Yb2Ti2O7 at 50 mK using polarised neutrons on D7 at the ILL. All the features are due to magnetism in the sample. This is a characteristic pattern from a Quantum Spin Liquid, stabilised by the presence of low levels of oxygen vacancies. The corresponding measurement from the same sample after the vacancies have been eliminated by annealing in oxygen exhibits no diffuse scattering.
Neutron scattering establishes that an exotic state of magnetism, a Quantum Spin Liquid, is stabilised by the presence of oxygen defects.
Magnetic materials are made of atoms that carry magnetic moments. Most magnetic materials will ‘order’ when they are cooled, meaning that the magnetic moments will arrange their directions to create a unique lattice. The precise layout of the lattice depends on the strength of the interactions between the magnetic atoms in the compound.
However, sometimes the material will not order. The positions of the magnetic atoms and the interactions between them are such that any order is ‘frustrated’. A magnetic moment on one atom may wish to orient itself in one direction with respect to another, but this puts it in direct competition with how it wishes to orient itself with respect to a third. The resulting ‘frustrated magnets’ are currently highly investigated, as they might host an array of fascinating physics. For example, some of these frustrated magnets display properties that have direct analogues in high-energy physics, such as magnetic monopoles and Majorana fermions, and a small crystal on a desktop may yield the same science as is studied at a huge particle collider like CERN.
Yb2Ti2O7 is an extremely interesting example of one such frustrated magnet. It shows characteristics of being a Quantum Spin Liquid, which is a state of matter in which the magnetic moments display some order over only a few atoms but not over the entire crystal. Furthermore, the moments never ‘lock in’ to fixed positions, but continue to fluctuate and move even at the lowest possible temperatures. The fluctuations pair up and entangle, encoding information across many magnetic moments. Good examples of Quantum Spin Liquid materials are highly sought after as they may have applications in quantum computing.
However, experimental studies on the compound have been contradictory. Measurements on powdered samples are consistent, showing a transition to some new state of matter at temperatures only a few thousandths of a degree above absolute zero. Measurements on single crystals are required to elucidate this state, which could be a ferromagnetic ordered state (via a so-called Higgs transition, which has an analogy in the search for the Higgs boson), a Quantum Spin Liquid, or a Quantum Spin Ice. The results of experiments on single crystals have not been consistent with those from powders, and have varied considerably between individual crystals.
Consequently, the exact state adopted at the lowest temperature by Yb2Ti2O7 is still unknown.
The state, and source of the contradiction, may now have been explained. A team led by Professor Jon Goff from Royal Holloway University, UK, has proposed that the discrepancies are due to crystal imperfections, or ‘disorder’, due to variations in the number and positions of the oxygen atoms in the crystals. They tested the hypothesis by employing neutron scattering techniques at the ISIS (UK) and ILL (France) facilities, measuring large crystals with oxygen disorder, and then measuring the same crystals after removing their disorder by heating them to 1200 degrees C in oxygen.
The experiments were supported by sophisticated Monte-Carlo calculations. The work, recently published in Nature Communications, established that defect-free Yb2Ti2O7 will form a type of ferromagnetic state at the lowest temperature. Introduce some oxygen defects, however, and the compound will form a Quantum Spin Liquid. It is the presence of defects that supplies the extra frustration required to create this new exotic state of matter.
“The high neutron flux of the ILL is ideally suited to the study of very weak diffuse scattering from disorder, and the ability to use polarised neutrons on D7 allows us to separate the structural and magnetic contributions" explains Professor Goff.
Dr Andrew Wildes, from ILL, says: "Let's face it: the Ytterbium Titanate controversy has been one of the biggest stories in recent studies on frustrated magnetism. Jon and his team have made an excellent effort to help solve the problem, and we were delighted to have been able to help.”
The work by Professor Goff leads to interesting possibilities. Can the oxygen defects be controllably manipulated in Yb2Ti2O7, so that the Quantum Spin Liquid characteristics could be tailored to measure? Are there other compounds whose exotic magnetic properties are similarly driven by disorder? Are there other compounds who may be induced to exotic magnetism through introducing disorder? Future work is planned!
ILL instrument: D7, diffuse scattering spectrometer
Re.: Nature Communications 2019; 10: 637. DOI: 10.1038/s41467-019-08598-z
Contact: Dr Andrew Wildes