Nanoscale Magnonic Neurons

TL;DR

This paper demonstrates the use of micromagnetic simulations to create nanoscale magnonic neurons using 2D chiral magnonic resonators, enabling potential development of magnonic neural networks.

Contribution

It introduces a novel design of nanoscale magnonic neurons based on nonlinear spin wave scattering in 2D resonators, advancing magnonic neural network technology.

Findings

Edge mode exhibits positive nonlinear frequency shift with increased spin wave amplitude.

Scattered spin waves are strong enough to activate secondary neurons.

Design enables connectivity for 2D magnonic neural networks.

Abstract

We use micromagnetic simulations to demonstrate neuron functionality of two-dimensional (2D) chiral magnonic resonators. Our design exploits nonlinear resonant scattering of spin waves propagating in a YIG medium from an edge mode of a permalloy nano-element. The reduced frequency and volume of the edge mode facilitate matching it to the YIG modes and give rise to their wide-angle scattering. As the amplitudes of the incident spin waves increase, the edge mode exhibits a positive nonlinear frequency shift. This shift leads to a complex frequency-dependent nonlinear variation of the amplitude and phase of spin waves scattered in different directions. We show that the scattered waves are strong enough to activate secondary neurons. This provides the connectivity required for combining our proposed neurons into 2D magnonic neural networks.