Eurocores_EuroGRAPHENE_Graphene-based systems for spintronics (SPINGRAPH)

Spintronics, or spin electronics, refers to the study of the role played by the electron spin in solid state physics, and the development of possible devices that specifically exploit spin properties instead of (or in addition to) charge degrees of freedom A special role in spintronics may be played by graphene, a single layer of hexagonally ordered carbon, with intriguing and unique properties, which is probably the best possible material for the realization of a metal-based transistor. Experimentally, graphene will be grown on top of Ni(111), Co(0001), and SiC(0001) substrates. Within the research consortium a range of well-established complementary experimental techniques is available in their groups to investigate electronic and magnetic properties of graphene-based systems such as scanning tunneling microscopy (STM), electron diffraction, angle- and spin-resolved photoemission as well as scanning tunneling spectroscopy (STS) using magnetic tips. The experimental investigations are augmented by the efforts of the present theory group, which has a strong scientific background in the determination of the geometrical structure, electronic structure and materials properties of the bulk, surfaces and interfaces of magnetic and non-magnetic materials using advanced highly precise DFT program packages developed in Vienna. The three central goals of the theory project are: To perform high quality DFT calculations of the spin-polarized electronic structure of mono and bilayers of graphene on and between Ni(111) and Co(111) surfaces at the corresponding optimal geometries and to use these results in post-DFT (GW) steps to calculate realistic quasi particle energies for the graphene bands. To model small graphene islands and narrow stripes on Ni(111) as well as small one and two layers thick Ni islands and narrow stripes on graphene/Ni(111), paying particular emphasis to edge states and their magnetic behavior. To study the electronic transport (conductivity) of suitably chosen graphene/Ni(111) model systems parallel and perpendicular to graphene layers in contact with a magnetic Ni surface, with particular emphasis on an expected spin filtering due to spin-dependent electronic transport.

Van-der-Waals (vdW) interactions are essential for a correct treatment of graphene (Gr) interface systems. Graphene consists of just one atomically thin, but chemically and thermally stable, single layer of carbon atoms strongly bound in a honeycomb arrangement. Its electronic properties (band-structure) are most intriguing since it is neither an insulator, like diamond nor a metal, like graphite, but should be seen as a zero-gap semiconductor. It features an unique property: charge carriers have zero effective mass resulting from a linear energy-momentum relation (band dispersion) of the relevant carbon pz electrons (? band). The basic idea was that when grown on ferromagnetic substrates such as the perfectly lattice matched nickel, the overlap of majority and minority charge carriers with the unique electronic states in graphene would cause a large spin anisotropy in electron transmission and thus might be used to create spin-dependent transport (spin filtering) in graphene-metal sandwich layers. The Austrian node provided theoretical support to the experimental groups in Berlin, Konstanz and Trieste, mostly aimed at a deeper understanding how the interaction between graphene and metal negatively influences the desired linear band dispersion. This was achieved by the application of both density functional theory (DFT) and post-DFT many-body methods for the determination of the atomic and electronic structure of the graphene/metal interface. These results were used to interpret experiments such as photoelectron spectroscopy and tunneling microscopy. The calculations revealed early on the absolute necessity to include van-der-Waals (vdW) interactions on top of the commonly used local functionals to describe the electronic interactions, or to systematically evaluate many-body interactions, which naturally include them. Calculations for Gr/Ni(111) showed that, despite adsorbing weakly, graphene interacts strongly with the Ni 3d-states and the graphene bands deviate heavily from linearity in the relevant regions of overlap, i.e. the desired transport properties are lost. The same holds true if one tries to retain the linear dispersion of the graphene bands in the weakly interacting Gr/Ir(111) system by intercalation of magnetic Ni atoms between graphene and Ir, thereby switching the interaction from weak to strong. The interaction may be also switched in the opposite way, for example by the intercalation of a weakly interacting material (Ag) in a strongly interacting Gr/Re(0001) interface. But even here the DFT calculations reveal that the graphene layer is still not completely decoupled but hybridizes with Ag d-states. The vdW-DFT results also reveal a significant influence of a metal substrate on the growth of a graphene sheet: the presence of a nickel interface stabilizes defects and lowers the energetic barriers to heal them. Also the presence of non-epitaxial rotated Gr/Ni(111) domains is explained on the basis of vdW-DFT calculations, which tie the rotated domains to a weakly coupled graphene layer on a nickel surface-carbide phase.