Understanding Spin-Transfer Torques using Ab-initio Simulations

• Applicant: Privatdozent Peter Elliott, Ph.D.
• Subject Area: Theoretical Condensed Matter Physics
• Term: from 2020 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 438494688

The overarching goal of this project is to use first-principles simulations tounderstand the microscopic physics of Spin-Orbit Torque, and how it can be controlled using femtosecond laser pulses.Spin-Orbit Torque (SOT) and the related Spin-Transfer Torque (STT) will likely underpin all future technology as they provide an efficient method to read and write data. Both SOT and STT make use of the electronic spin-current, rather than the charge current, to create torque on local magnetic moments, the direction of which encodes the data. Thus, by understanding the microscopic, quantum-mechanical, physics of these processes, we can design better devices to meet our future technological needs. However, the physics of these phenomena is not yet completely understood.In such a situation, it becomes useful, if not imperative, to perform ab-initio simulations. These first-principles simulations make no assumptions on the underlying physics and simply calculate the response of a material to external electric and magnetic fields. Thus, I will simulate the behavior of the STT and SOT in different situations (e.g. changing the materials and geometries) to isolate and elucidate the various physical processes involved. Such calculations can be computationally demanding as they must solve the Schrödinger equation of quantum mechanics for many interacting electrons. To remedy this problem, I will use the computationally efficient, but still ab-initio, method of time-dependent density functional theory (TDDFT). In the past 5 years, our research group has proven that TDDFT can successfully explain and predict several interesting phenomena in ultrafast spin dynamics. Thus, it is an effective tool to use in this study.The spin currents required for STT/SOT can be generated by inducing a current in a magnetic material or by using the Spin Hall effect (SHE) in non-magnetic materials. Both mechanisms will be studied in this project. Furthermore, the SHE is very interesting in its own right, particularly in the ultrafast regime where many current SHE models are inapplicable. For this reason, I will freely distribute (under the Gnu public license) the code I develop and implement to calculate the spin-Hall effect in the linear response and non-linear regimes.Finally, I will study how the STT/SOT behaves on ultrafast timescales by applying femtosecond laser pulses. These timescales are approximately a million times faster than devices currently available. By studying how the spin-currents and magnetic torques depend on the laser parameters (e.g. frequency, duration, intensity), I can understand how STT/SOT can be advanced into this new domain.

• DFG Programme: Research Grants