News Release

Riddle of Kondo effect solved in ultimately thin wires

Peer-Reviewed Publication

University of Cologne

A team of physicists at the University of Cologne has solved a long-standing problem of condensed matter physics: they have directly observed the Kondo effect (the re-grouping of electrons in a metal caused by magnetic impurities) visible in a single artificial atom. This has not been done successfully in the past, since the magnetic orbitals of atoms usually cannot be directly observed with most measurement techniques. However, the international research team led by Dr Wouter Jolie at the University of Cologne’s Institute for Experimental Physics used a new technique to observe the Kondo effect in an artificial orbital inside a one-dimensional wire floating above a metallic sheet of graphene. They report their discovery in the article “Modulated Kondo screening along magnetic mirror twin boundaries in monolayer MoS2” in Nature Physics.
When electrons moving through a metal encounter a magnetic atom, they are affected by the atom’s spin – the magnetic pole of elementary particles. In trying to screen the effect of the atomic spin, the electron sea groups together close to the atom, forming a new many-body state which is called the Kondo resonance. This collective behaviour is known as the Kondo effect and is often used to describe metals interacting with magnetic atoms. However, other types of interactions can lead to very similar experimental signatures, questioning the role of the Kondo effect for single magnetic atoms on surfaces.
The physicists used a new experimental approach to show that their one-dimensional wires are also subject to the Kondo effect: the electrons trapped in the wires form standing waves, which can be thought as extended atomic orbitals. This artificial orbital, its coupling to the electron sea, as well as the resonant transitions between orbital and sea can be imaged with the scanning tunnelling microscope. This experimental technique uses a sharp metallic needle to measure electrons with atomic resolution. This has allowed the team to measure the Kondo effect with unparalleled precision. 
“With magnetic atoms on surfaces, it is like with the story about the person who has never seen an elephant and tries to imagine its shape by touching it once in a dark room. If you only feel the trunk, you imagine a completely different animal than if you are touching the side,” said Camiel van Efferen, the doctoral student who conducted the experiments. “For a long time, only the Kondo resonance was measured. But there could be other explanations for the signals observed in these measurements, just like the elephant’s trunk could also be a snake.”
The research group at the Institute of Experimental Physics specializes in the growth and exploration of 2D materials - crystalline solids consisting of just a few layers of atoms – such as graphene and monolayer molybdenum disulfide (MoS2). They found that at the interface of two MoS2 crystals, one of which is the mirror image of the other, a metallic wire of atoms forms. With their scanning tunnelling microscope, they could simultaneously measure the magnetic states and the Kondo resonance, at an astonishingly low temperature of -272.75 degrees C (0.4 Kelvin), at which the Kondo effect emerges.
“While our measurement left no doubts that we observed the Kondo effect, we did not yet know how well our unconventional approach could be compared to theoretical predictions,” Jolie added. For that, the team enlisted the help of two theoretical physicists, Professor Dr Achim Rosch from the University of Cologne and Dr Theo Costi from Forschungszentrum Jülich, both world-renowned experts in the field of Kondo physics. After crunching the experimental data in the supercomputer in Jülich, it turned out that the Kondo resonance could be exactly predicted from the shape of the artificial orbitals in the magnetic wires, validating a decades-old prediction from one of the founding fathers of condensed matter physics, Philip W. Anderson.
The scientists are now planning to use their magnetic wires to investigate even more exotic phenomena. “Placing our 1D wires on a superconductor or on a quantum spin-liquid, we could create many-body states emerging from other quasiparticles than electrons,” explained Camiel van Efferen. “The fascinating states of matter that arise from these interactions can now be seen clearly, which will allow us to understand them on a completely new level.” 

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.