Spin currents and spin transfer torque produced by Spin Orbit effects – SOspin

Spintronics is based on the control of spin currents to carry and manipulate information. For instance, Giant and Tunnel Magnetoresistances, phenomena at the heart of several spintronics such as hard disk heads or MRAM memories, rely on how spin currents flow in magnetic heterostructures. Beyond magnetoresistance effects, spin currents can also produce spin transfer torques, and thus be used to manipulate magnetization states using solely electrical currents. Nowadays, the race towards the industrial developments of devices based on the spin transfer torque is on, in particular with mature R&D programs on the STT-RAM, a spin transfer torque-based MRAM (Samsung, IBM, Hitachi,...).

The SOspin project aims at studying and maximizing the effects related to the spin-orbit coupling and allowing to generate and to detect spin currents. The main idea is to control the different spin current sources, i.e., the volume effects (spin Hall Effect or SHE) and the interface effect (Rashba effect). These developments should allow realizing innovative spin-current generators, particularly to induce spin-transfer torques.

We will firstly study the possibility to increase the SHE by using recent results obtained in an Au-based alloy. We showed that when the SHE is dominated by the side-jump effect, the SHE angle (i.e., the efficiency of the SHE) increases with the impurity concentration. The main idea is to increase the impurity concentration up to the solubility limit, in order to maximize the SHE angle.

We will also study interface effects as the Rashba effect. Preliminary results show that the Ag/Bi system exhibits a large conversion of spin current into charge current. This effect will also be studied in Ag/Au. Its theoretical understanding will also allow us to study Ta/Co and Pt/Co-like systems, with an interface between a non-magnetic and a ferromagnetic metal. Indeed, although very large spin-transfer torque effects have been observed in these systems, the origin of their efficiency could reside either in volume (SHE) or interface (Rashba) effects, this question generating a major debate in the spintronics community.

Finally, we will try to combine both the volume and interface effects in order to maximize the spin current production and the spin transfer torque. To realize this project, we will use complementary measurement techniques, both static and dynamic, in order to characterize the produced spin currents and spin transfer torques. We will also realize theoretical developments necessary to the understanding of these effects.