Neutron-scattering experiments have revealed an important clue as to why a recently discovered family of iron-based materials are superconducting at relatively high temperatures. The measurements were made by researchers in the US and UK and show that the "superconducting energy gap" in such materials is different from that found in either conventional or cuprate superconductors. This suggests that a new mechanism of superconductivity is at work.

Superconductivity occurs when a material is cooled below a certain temperature and its conduction electrons form pairs that can flow without any resistance. The superconducting energy gap, which is an important physical property of a superconductor, is the energy required to break one of these pairs. Indeed, measuring the precise nature of the gap is crucial for understanding the physics of superconductivity.

In a conventional, low-temperature superconductor, such as lead, physicists know that the gap is perfectly symmetrical with respect to the direction of the momentum of the electrons. Such superconductors can be described using BCS theory, developed in 1957 by John Bardeen, Leon Cooper and Robert Schrieffer. The theory shows that electron pairs are created as a result of interactions between electrons and vibrating atoms in the material.

However, in other materials, such as the high-temperature (high-Tc) cuprate superconductors, the gap is not perfectly symmetric, but rather has distinct lobes described as “d-wave” symmetry. As BCS theory requires a symmetric gap, it therefore cannot be used to explain the behaviour of high-temperature superconductors. While there seems to be a strong relationship between gaps with unusual symmetry and high-Tc superconductivity, physicists are still trying to understand the pairing mechanism in such materials.

Symmetric or not?

To complicate matters even further, the new family of iron-arsenide-based high-Tc superconductors that physicists discovered last year do not appear to fit either the BCS or cuprate models. When the energy gaps of these materials are measured using the standard technique of angle-resolved photoemission spectroscopy (ARPES), the gaps appear to be symmetric — but BCS theory is unable to explain why they are superconductors at relatively high temperatures.

Some physicists have realized that it could be possible for a superconductor to have a gap that is perfectly symmetric in terms of its magnitude, but with a positive phase for some electrons and a negative phase for others. A gap with such "S±" symmetry could be related to iron arsenide’s high-Tc behaviour.

Now, Ray Osborn and colleagues at Argonne and Oak Ridge National Labs and Northwestern University in the US and the Rutherford Appleton Lab (RAL) in the UK have used inelastic neutron scattering to find the first experimental evidence for S± symmetry in an iron arsenide superconductor containing some barium and potassium (Ba0.6K 0.4 Fe 2As2).

Merlin reveals magnetic excitation

The team used the Merlin spectrometer at the ISIS neutron scattering facility at the Rutherford Appleton Laboratory in the UK (Nature 456 930). Pulses of neutrons are fired at the material and the energy and momentum distributions of the scattered neutrons are measured. The team found that the neutrons were causing a magnetic excitation in the material at a unique momentum and energy — which they believe can only occur if the energy gap has S± symmetry. According to Osborn, this symmetry had not been detected before because ARPES is not sensitive to the phase of the gap.

Osborn told physicsworld.com that the S± symmetry has “profound consequences for the nature of the pairing interaction”. Unlike the BCS interaction, which is attractive over short distances, Osborn says that S± symmetry implies that the pairing mechanism in their superconductor is repulsive at short distances and attractive at longer distances. “That’s important information for theorists”, he said, but cautioned: “it doesn’t say what the precise mechanism actually is”.

According to Igor Mazin at the Naval Research Lab in Washington, the discovery means that it is very likely that the mechanism causing superconductivity in iron-arsenide high–Tc materials is different from that in either the cuprates or the conventional BCS materials.

Mazin also pointed out that the S± symmetry also means that it is likely that magnetism is involved in the pairing interaction and that a better understanding of the iron arsenide materials will be the key to understanding their superconductivity.

Osborn and colleagues are already planning to do further neutron scattering experiments on single-crystal (rather than polycrystalline) samples, which should reveal more information about the S± gap.