Magnetic order in a coherent two-dimensional Kondo lattice

While mounting evidence exists that transition metal dichalcogenide heterobilayers realize a novel 2D Kondo lattice of effective magnetic moments, their low temperature ground state remains to be established. In this talk I will present our scanning tunneling spectroscopy data for 1T/1H-TaSe2, showing the existence of a split Kondo peak around the Fermi energy, spatially located at the central Ta atom of the charge density wave of the 1T phase. The splitting of this peak further increases with a mangetic field in a non-linear fashion. Aided by ab-initio calculations, I will show that this system is in the limit of small Kondo echange where magnetic order is expected in the Doniach phase diagram. I will then discuss how the non-linear splitting of the peak is also inconsistent with a fully screened Kondo lattice, and originates instead from a ground state with magnetic order, as expected from the strength of the coupling. I will end with a general discussion of the phase diagram of this system.