Self-assembly of correlated Kondo lattices: The Mott to Kondo transition in diluted superlattices

Abstract

In the field of condensed matter, the quest to obtain an experimental realization of a Kondo lattice has generated a tremendous effort of the community, from both standpoints of experiments and theory. The pursuit of obtaining independent magnetic moments, via charge localization through Coulomb interactions, is paramount for applications in nanotechnology. In particular, systems with simultaneous charge and spin degrees of freedom can manifest both Kondo spin quenching and Mott-Hubbard charge localization. A unified experimental framework illuminating the pathway between the two phenomena is of physical and technological interest, and is (as of yet) hardly observed in real condensed matter systems. Recent developments in the ability to control densities and temperatures of strongly correlated Fermionic impurities on surfaces and substrates has opened up a new paradigm of possibilities for this pathway. In particular, a milestone was recently surpassed through the observation of self-assembled superlattices of f band adatoms on metallic surfaces, such as the deposition of Ce on Ag(111). Such lattices have introduced a mechanism of diluted correlated lattices where the interaction between Kondo and Mott physics can be methodically studied. However, it remains difficult to control the adatom distances and substrate densities in these systems, and the interplay between Kondo physics and charge localization remains elusive. In this work, we systematically investigate the phase diagram of superlattice structures of heavy f elements deposited on metallic substrates, and assess the required conditions to obtain Kondo lattices in superlattices. We unveil a unique pathway between Kondo quenching and Mott localization, and identify a non-trivial charge density wave phase emerging from the competition of charge localization and Kondo physics.