Spin transport in graphene: Fundamental concepts and practical implications

Electronic devices based on spin transport are expected to play a major role in future Information and Communication Technologies (ICT). Those spintronics devices will use the spin degree of freedom to store, transport and process information. Even if magnetic read heads have been the cornerstone of magnetic hard drives since 1996, it is only recently that the full potential of spintronics has been included in the ITRS road map. Spin information processing will require the ability to inject, manipulate and detect spins. To this respect, the seminal spintransistor proposed by Datta and Das [1] is a very simple device based on spin injection and detection using ferromagnetic electrodes and on the manipulation of the spin information through the spin-orbit coupling (Rashba effect). It was, however, soon understood that fundamental constraints on the physics governing spin transport will make this concept very difficult to achieve with conventional semiconductors (GaAs or Si) [2, 3]. Here we will present a set of results on an alternative route where the channel is no longer a conventional semiconductor but graphene. Graphene is expected to be a good candidate for spin information transport as its mobility at room temperature outperforms that of any other material [5] and its spin orbit coupling as well as the hyperfine interaction of ¹²C are expected to be very small. Using Multilayer graphene grown on SiC wafers we experimentally demonstrate using spin transport experiments in the lateral spin-valve geometry that the spin diffusion length is at least of the order of 100 μm [6].