Magnetic Impurities in SuperconducToRs: from single Atoms to Lattices – MISTRAL

We have studied superconducting crystals with a lamellar structure, giving to the electronic structure a quasi- two-dimensional behavior. The growth of these monocrystals was achieved by adding a small percentage of iron impurities leading to the inclusion of magnetic defects homogeneously distributed in the sample. By moving the microscope tip over the sample, we were then able to measure the spatial dependence of the tunnel current and reconstruct the distribution of electronic states and thus image the structure of the Yu-Shiba-Rusinov state. The results obtained by scanning tunneling microscopy are compared with the theoretical numerical calculations in terms of the spatial extension of the wave function, of its oscillating nature and of its star shape structure.

During the first 18 months of this proposal, we have revisited in details the interaction between a single magnetic atom encapsulated in a superconducting matrix (here NbSe2 ) paying attention to the spatial dependence and extension of the Shiba bound states which have been characterized spectroscopically by scanning tunneling spectroscopy. Such real-space cartography of the Shiba bound states is one the most novel aspect of our recent result. We have been able to reach it due to our clean sample preparation and the high energy and spatial resolution of our mapping of the local density of states around magnetic impurities. By analyzing different magnetic impurities in the layered superconductor NbSe2. We have found that the spatial extension of the Shiba bound states is much larger than reported in the very few works published earlier. Instead, we experimentally found that the spatial extent can be of order of 10nm which is 20 times larger than the spatial extent found earlier. By directly solving numerically the Shiba BS equations self-consistently for an arbitrary band Hamiltonian (we used a tight binding faithful description of the band structure of NbSe2), we have recovered the typical flower shape structure observed experimentally together with the measured spatial extent. Most importantly, by taking a continuum limit description of the Shiba equations that allows extracting analytically the two lengths scales shaping the local density of states around the impurity, we have been able to show unambiguously that this large spatial extent is due to the 2D character of the layered superconductor.

We now want to extend this analysis to magnetic clusters in 2D superconductors.

Gerbold C. Ménard, Sébastien Guissart, Christophe Brun, Stéphane Pons, Vasily S. Stolyarov, François Debontridder, Matthieu V. Leclerc, Etienne Janod, Laurent Cario, Dimitri Roditchev, Pascal Simon, Tristan Cren, Nature Physics 11,1013–1016 (2015).

A single classical spin immersed in a superconductor induces the so-called Shiba bound states within the superconducting gap. A finite concentration of magnetic impurities can form an impurity band from the overlapping Shiba states. It was recently predicted that a chain of magnetic impurities on the surface of a superconductor may realize a one-dimensional topological superconductor hosting Majorana end states. The main purpose of this proposal is to provide a systematic bottom up experimental and theoretical analysis of such a complex system. We first plan to develop a better understanding of the interaction between a single magnetic atom and a superconductor. A particular attention will be paid to the spatial dependence of the bound states and their spectral signature as measured by scanning tunneling spectroscopy. We will then address the crucial issue of the magnetic interaction between two spins separated by a distance r before going toward the limit of ordered arrays of magnetic impurities.