Past Events

Abstract
 

As a collective quasiparticle excitation of the magnetic order, spin wave can propagate in both conducting and insulating materials with little dissipation. Magnetic material with non-trivia magnetic texture provides a unified information memory and processing platform [1, 2].

We demonstrate that, in the presence of Dzyaloshinskii-Moriya interaction, an antiferromagnetic domain wall acts as a spin wave polarizer or a spin wave retarder [3]. Based on the same principle, the spin wave driven domain wall motion in antiferromagnet also strongly depends on the linear polarization of the injected spin waves [4].

Furthermore, we construct functional spin wave logic devices, in which information processing is realized by manipulating spin wave polarizations. This magnonic logic gate has several advantages comparing to existing logic architectures: i) high speed, reaching the extremely high frequency of antiferromagnet (THz); ii) energy efficient, the insulator based spin wave computing can avoid electronic Joule heating completely; iii) simple design, both logic and memory parts base on a common magnetic material/system thus reducing the complexity in hardware. Following the demonstration of the functionality of individual gates, we present a function halfadder composed of two gates. [4]

The proposed theoretical scheme for a spin wave computing is intrinsically a non-von Neumann architecture with processing-in-memory.

Anyone interested is welcome to attend.