Spin electronics, or spintronics, relies on the spin of the electrons, rather than their charge, to transport, manipulate and store information in an electronic device. Modern spintronic technologies, including magnetic sensors and magnetic memories, rely on the non-volatility (storage of information) provided by ferromagnetic materials. In order to unlock its full potential, spintronics still requires a suitable template to transport and manipulate the spins, which would enable the implementation of novel spin-logic architectures with very low power requirements. Graphene and engineered graphene are amongst the most promising candidates to fill this gap. Spins are expected to be conserved over long distances in pristine graphene, and could in principle be manipulated within graphene regions that are modified by the proximity of a ferromagnetic insulator or a material with large spin-orbit interaction.
Such advances require full understanding and control of the behaviour of the spins in graphene. Nevertheless, after 10 years of intense research, even the basic process leading to the loss of spin information in pristine graphene remains largely debated. This is a fascinating puzzle rooted in the properties of this unique material. Indeed, graphene is now believed to support a number of spin relaxation mechanisms with no equivalent in any previously studied system, even though these mechanisms have yet to be established experimentally. We have developed and demonstrated a novel approach to solve this puzzle based on the determination of the spin relaxation anisotropy. Graphene is a two-dimensional system and the spin relaxation anisotropy quantifies the difference between the relaxation rates of spins oriented in- or out-of- the plane of graphene. We show that its magnitude provides direct evidence of the spin relaxation mechanisms at play. Future work will focus on modifying graphene to achieve full control of its spin dynamics properties.