The spin Hall effect - searching for ideal materials

Martin Gradhand

University of Bristol, Faculty of Science, School of Physics, Bristol, UK

During the last years, it became possible to measure electronically the spin Hall effect (SHE) in metallic devices. It opens the opportunity to use it for spin current generation in metal-based spintronics. This avoids the difficulties connected with spin current injection from ferromagnets into semiconductors. With this perspective materials are requested which provide a strong effect, i.e. which show an effective conversion of charge currents into spin currents.
The origin of the SHE is the spin orbit coupling (SOC). It is often expected that strong SOC leads automatically to large SHE, which is not generally true. Furthermore the spin diffusion length decreases with increasing SOC. Since the spin diffusion length restricts the distance over which the spin information is maintained it limits the dimension of a device in potential applications.
Here we present an ab initio analysis for dilute alloys to optimize their potential for an effective charge to spin current conversion in combination with a long spin diffusion length.

We considered the extrinsic mechanism by calculating the skew-scattering event at impurities in an ideal lattice. This input from our ab initio code, a Korringa-Kohn-Rostoker Greens function method, is used to solve a linearized Boltzmann equation.

The intrinsic contribution was calculated via the Berry curvature description of the anomalous velocity. For a new implementation in our code we used special features of the screened KKR method for an efficient calculation of the Berry curvature.

Combining all tools we were able to identify systems showing a charge to spin current conversion of nearly 10% in combination with a spin diffusion length as large as 100nm.

Back