Distinct, but not so different

Among superconducting materials, CeCoIn5 stands out as a rare case where superconductivity gives rise to magnetic order. An international team led by PSI physicist Michel Kenzelmann now reports that when small amounts of impurities are implanted into CeCoIn5, then two distinct magnetic phases appear — and these are surprisingly similar to one another.

The interactions between electrons in a solid can give rise to a host of intriguing collective behaviours, most notably to magnetic order and to superconductivity. It has long been known that in some materials these two phenomena can co-exist, but typically they are competing — as one emerges, the other is suppressed.

Cooperation rather than competition

In 2008, Kenzelmann and colleagues showed that in the compound CeCoIn5, superconducting and magnetic orders cooperate rather than compete [1]. Indeed, magnetic order exists only when the material is cold enough and in a sufficiently low magnetic field to be superconducting. Further insight into the intricate interplay between superconducting and magnetic order in this material came in 2014, when a team led by the PSI physicist proposed that at high fields superconductivity itself carries a magnetic moment in CeCoIn5 and is spatially inhomogeneous [2]. This observation supports the notion of a so-called pair-density wave — pertaining to the superconducting state — that co-exists with a spin-density wave (SDW), a form of magnetic order. Turning now to a slightly modified version of CeCoIn5, they found two distinct phases with SDW order, separated by a possible quantum phase transition [3].

Surprising re-appearance act
 

A magnetic instability was found in Nd0.05Ce0.95CoIn5 that separates two magnetic phases with identical symmetry inside the superconducting condensate (taken from ref. 3).

For this work, Kenzelmann’s team studied a crystal in which 5 % of the cerium (Ce) atoms in CeCoIn5 have been substituted by neodymium (Nd). From other studies, they knew that SDW order exists in such a doped compound. But how the Nd impurities influence the overall dynamics of the electrons in the material remained unclear. Daniel Mazzone, a PhD student in Kenzelmann’s group at PSI, together with co-workers in France and Denmark, used neutron diffraction to further explore magnetic order in Nd0.05Ce0.95CoIn5. A magnetic Bragg peak — a tell-tale sign for magnetic order — was clearly present at relatively low magnetic fields. However, as they increased the field to 8 Tesla, that peak vanished, only to appear again at even higher fields. Finally, at 11 Tesla the magnetic order collapsed.

The collapse at 11 Tesla was expected, as there superconductivity breaks down. Surprising, however, was the ‘magnetic instability’ separating the superconducting phase into two distinct regions. The symmetries of the SDWs detected in the low-field and the high-field magnetic phase, respectively, were found to be identical. This means that the transition from one to the other is not driven by magnetic fluctuations. Instead, the findings of Mazzone et al. point towards a quantum phase transition, an interpretation that is also consistent with the 2014 observations regarding qualitative changes in the superconducting behaviour in undoped CeCoIn5.

A direct proof for the existence of a quantum phase transition is yet to be given. Nonetheless, these new results add substantially to an ever-fuller understanding not only of CeCoIn5, a material from which exceptionally clean samples can be produced, but also of other ‘exotic superconductors’ where we still lack a microscopic understanding of their electronic properties.