Xiao Lab publishes Nature Communications article on the emerging field of spin-orbit torques

May 1, 2013 - The recent Nature Communications article by Prof. Xiao and his graduate students and postdocs, including also a collaborator from the State Key Laboratory of Electronic Films and Integrated Devices in China, has provides perhaps the key insight into the controversy surrounding recently emerged fields of spin torques driven by spin-orbit (SO) coupling generated by relativstic effects in solids. 

The spin torque is a phenomenon in which a spin current of sufficiently large density injected into a ferromagnetic (F) layer either switches its magnetization from one static configuration to another or generates a dynamical situation with steady-state precessing magnetization. The origin of STT is the absorption of the itinerant flow of angular momentum components normal to the magnetization direction. It represents one of the central phenomena of the second-generation spintronics, focused on manipulation of coherent spin states, since the reduction of current densities (currently of the order 106-108 A/cm2) required for STT-based magnetization switching is expected to bring about commercially viable ST-magnetic-random-access-memories. The rich nonequilibrium physics arising in the interplay of spin currents carried by fast conduction electrons and slow collective magnetization dynamics is also of great fundamental interest.

Unlike conventional spin torque observed in F/I/F magnetic tunnel junctions (I-insulating barrier), where two F layers are required with their magnetizations being non-collinear, the recently discovered spin torques in F/N bilayers involve only one F layer, as well as strong SO coupling at the F/N  interface or in the bulk of the N layer. Passing current parallel to the F/N interface can generate such unconventional torques either due to nonequilibrium spin accumulation at the interface or spin Hall current generated within the N layer which then flows into the F layer. However, various experiments have reported conflicting results for the size and direction of the torque, whose sensitivity on the thickness of N layer is difficult to reconcile with theoretical predictions.

The article by Xiao Lab demonstrated sensitive SO effective field measurements up to 10 nm thick F layer to find how the effective field rapidly diminishes with the increase of F layer thickness. They further showed that this effective field persists even with the insertion of a copper spacer. Thus, such nonlocal measurement suggests that the SO effective field does not rely on the heavy normal metal/ferromagnetic metal interface. The absolute current applied in-plane through a thin lateral surface of recently F/N bilayers could also provide significant gains in terms of integration of spin-orbit torques and power consumption. More information about spintronics projects pursued by the Xiao Lab is availabe from their Website.