Cuprates and pseudo-gap

Cuprates show some unusual magnetism above their superconducting temperatures, which may be key to their exotic electronic properties.

It is 30 years since scientists discovered a family of metal oxides, the cuprates, that can conduct electricity without resistance up to temperatures much higher (138K) than those associated with metallic superconductors (maximum temperature, 30 K).  However, even now, how the cuprates do it is still a subject of much debate and extensive research. The mechanism appears to be a good deal more complicated than in conventional superconductors, and is requiring new descriptions that extend the frontiers of fundamental physics.

One of the first high-temperature superconductors discovered was yttrium barium cuprate, YBa2Cu3O7, in which the oxygen content is lowered fractionally to create positive charges, or ‘holes’ in the crystal lattice that generate the necessary electron mobility. The level of such ‘hole-doping’ mediates the electronic (and magnetic) properties. To try to pin down what exactly goes on, physicists have mapped these properties over temperature and for different doping levels (see phase diagram opposite).

What the phase diagram reveals is that just above the temperature at which cuprates stop becoming superconductors, there is a strange region called a pseudogap, in which the material acts like a ‘bad metal’, behaving neither as a metal nor an insulator. This is quite different from low-temperature superconductors, which revert to a normal metals above the transition temperature. This pseudogap is turning out to be the Rosetta Stone for discovering the physical principles that underlie the cuprates’ complex behaviour, and so it is being extensively studied.

One aspect, becoming clear, is that subtle magnetic interactions found within the square planar arrangements of copper and oxygen ions (CuO2) that form layers in the cuprate crystal lattice are very significant (figure). Polarised neutron scattering (pxx) offers a way of studying these interactions because it can distinguish between normal and magnetic scattering.

Circulating loop currents

Such experiments have been used to look at the magnetic behaviour in the pseudogap zone in four cuprate families, including YBa2Cu3O6+x,. They have uncovered a curious type of antiferromagnetic order (that is, with ordered antiparallel magnetic moments, pxx), which operates within each unit cell of the crystal, and whose appearance coincides with the onset of the pseudogap. This magnetic order could be the result of loop currents flowing within the unit cells, generating staggered static orbital magnetic moments (figure). The loop currents appear in pairs flowing clockwise and anti-clockwise, and the order is long-ranged at low doping levels but vanishes at high doping levels. 

The existence of this intra-unit-cell magnetic order has now been well-documented over a wide hole-doping range. However, what happens to the proposed loop currents in compounds with optimal doping (where the superconducting ‘dome’ seen in the phase diagram shows its maximum) had not yet been addressed.

Our studies on optimally doped YBa2Cu3O6.85 showed that the intra-unit-cell magnetic order persists to relatively high temperatures. However, the magnetic intensity is strongly reduced compared to that in lower-doped samples, and the magnetic order is short-ranged. In addition, we confirmed previous reports that the orbital moment does not just point vertically to the copper–oxygen planes, as expected from the loop-current model, but at low temperatures is tilted resulting in an additional in-plane component. This may be due to the existence of loop patterns circulating in different directions and forming a superposition of quantum states – which could exist only at low temperatures. Whatever the reason, the discovery of these two sets of behaviours puts stringent constraints on any theoretical description of the pseudo-gap and thus high-temperature superconductivity.

Research team: L. Mangin-Thro, Y. Sidis and P. Bourges (LLB, CEA Saclay, France), and A. Wildes (ILL)

Instrument:D7 - Diffuse Scattering Spectrometer

Re.: Intra-unit-cell magnetic correlations near optimal doping in YBa2Cu3O6.85. Intra-unit-cell magnetic correlations near optimal doping in YBa2Cu3O6.85. L. Mangin-Thro, Y. Sidis, A. Wildes & P. Bourges. Nature Communications 6, Article number: 7705 (2015) doi:10.1038/ncomms8705

An excerpt from ILL topical brochure on Neutrons and Fundamental Physics, p 10