Graphene-mediated exchange coupling between Co single spins and Ni substrates

*Valerio Bellini (1), Alessandro Barla (2), Carlo Carbone (2), Stefan Heinze (3)
(1) Istituto Nanoscienze - CNR Nano S3, via G. Campi 231/A, 41125 Modena, Italy, (2) Istituto di Struttura della Materia (ISM), Consiglio Nazionale delle Ricerche (CNR), I-34149 Trieste, Italy, (3) Institute of Theoretical Physics and Astrophysics, University of Kiel, D-24098 Kiel, Germany.

We present a density-functional theory study of the exchange coupling between molecular/single atom spin systems and a magnetic substrate across a graphene spacer. We concentrate on cobaltocene (CoCp2) molecules and Co adatoms adsorbed on graphene decorated Ni(111) substrates. The role of graphene is to preserve the adsorbate magnetic moments, acting as an electronic decoupling layer, while allowing effective spin communication between the adsorbate and the substrate. In fact, sizable magnetic coupling is predicted by the theory, and in some fortunate cases can be also rationalized in term of spatial and energy matching between adsorbate and substrate orbitals.

This is the case of cobaltocene, where in virtue of the peculiar magnetic properties of the molecule, the coupling mechanism is more accessible, and the description of the employed theoretical methodology suffices to rationalize it, and even tailor it by, for instance, intercalation of different ferromagnetic metal monolayers, such as Fe and Co, between graphene and the Ni substrate [S. Marocchi et al., Phys. Rev. B. 88, 144407 (2013)].

Concerning single atom spin adsorbates, we first concentrate on Co adatoms, where a thorough experimental analysis by STM and XMCD experiments has been also conducted, and where a consistent description between theory and experiments can be found only by considering the presence of more adsorption sites characterized by different magnetic properties [A. Barla et al., ACS Nano 10, 1101 (2016)]. If we extend the analysis to the other elements of the 3d series, i.e. from Sc to Fe, although some trends across the series could be found [V. Bellini et al., to be submitted], no clear understanding of the magnetic coupling can be set forward. Moreover, the employed methodology fails in the comparison with XMCD experiments performed on V adatoms, sign that more refined multi-reference methods might be necessary to correctly describe the ground state magnetic properties of these systems.