Burkard Hillebrands: Magnon supercurrents

Abstract

With the fast growth in the volume of information being processed, researchers are charged with the primary task of finding new ways for fast and efficient processing and transfer of data. Spin excitations – spin waves and their quanta magnons – open up a very promising branch of high-speed and low-power information processing. In contrast to conventional spintronics, which relies on the transport of spin-polarized electron currents, magnon currents are spin fluxes, which propagate entirely without charge transfer through magnetically ordered, conducting or insulating, materials. By including elements, which convert information coded into spin or charge of electrons into magnons, magnon based spintronic systems combine the advantages of conventional spintronics with those of a uniquely versatile wave-based information platform. An extraordinary challenge is the use of macroscopic quantum phenomena such as magnon Bose-Einstein condensates (BEC) for the information transfer and processing.

Magnons are bosons, and thus they are able to form spontaneously a spatially extended, coherent ground state, which can be established independently of the magnon excitation mechanism even at room temperature [1]. In a classic condensate, the group velocity is zero. Recently we have succeeded to create a condensate with non-zero group velocity by spin-orbit coupling to phononic states. The condensate is formed in a parametrically driven, single-crystal film of yttrium iron garnet (Y3Fe5O12, YIG) [2].

Even more promising is use of magnon supercurrents, which constitute the transport of angular momentum, driven by a phase gradient in the magnon-condensate wave function. The dynamics of the magnon BEC in a thermal gradient, which was revealed by means of time- and wavevector- resolved Brillouin light scattering (BLS) spectroscopy, presents the first evidence of the formation of a magnon supercurrent [3].

The work was financially supported by the DFG within the SFB/TR 49.