Synaptic element for neuromorphic computing using a magnetic domain wall device with synthetic pinning sites

Abstract

The ability to make devices that mimic the human brain has been a subject of great interest in scientific research in recent years. Current artificial intelligence algorithms are primarily executed on the von Neumann hardware. This causes a bottleneck in processing speeds and is not energy efficient. In this work, we have demonstrated a synaptic element based on a magnetic domain wall device. The domain wall motion was controlled with the use of synthetic pinning sites, which were introduced by Boron (B+) ion-implantation for local modification of the magnetic properties. The magnetization switching process of a Co/Pd multilayer structure with perpendicular magnetic anisotropy was observed by using MagVision Kerr microscopy system. The B+ implantation depth was controlled by varying the thickness of a Ta overcoat layer. The Kerr microscopy results correlate with the electrical measurements of the wire which show multiple resistive states. The control of the domain wall motion with the synthetic pinning sites is demonstrated to be a reliable technique for neuromorphic applications.

Emulation of Synaptic Plasticity on a Cobalt-Based Synaptic Transistor for Neuromorphic Computing

“…To achieve these goals, several two-terminal synaptic devices have been widely investigated, especially by exploiting the functionalities such as phase-change memory, 4,5 atomic switch, 6,7 spintronics devices, 8,9 and memristors. 10−13 In twoterminal devices, the learning is usually achieved by feedback from the post neuron, and the signal transmission is highly depressed during the learning operation.…”

“…To achieve these goals, several two-terminal synaptic devices have been widely investigated, especially by exploiting the functionalities such as phase-change memory, , atomic switch, , spintronics devices, , and memristors. − In two-terminal devices, the learning is usually achieved by feedback from the post neuron, and the signal transmission is highly depressed during the learning operation. , Therefore, complete emulation of the synapse is limited as learning and signal transmission could not be performed simultaneously. Compared to two-terminal devices, in three-terminal synaptic devices, the learning operation and signal transmission process are realized simultaneously through the gate terminal and channel, respectively, resulting in complete emulation of a synapse.…”

Multi-Level Neuromorphic Devices Built on Emerging Ferroic Materials: A Review

“…For example, most DWM devices relies on precise control and sensing of a single domain wall, which often needs a quasi-1D device shape for confinement of the domain dynamics. Such constraint on device geometry lead to large footprints along the DW propagation direction (ranging from 1 to 10 μm), and thus hindering the deployment of the DWM devices beyond prototype demos (Cai et al, 2017 ; Jin et al, 2019 ; Siddiqui et al, 2019 ). Novel ideas such as introducing skyirmions have been investigated (Chen et al, 2018 ), potentially extending the current single wall based devices to multiple DW configuration (Song et al, 2020 ), but more work is needed to illustrate a viable path of scalability and controllability using skyrmions.…”

Domain wall pinning through nanoscale interfacial Dzyaloshinskii–Moriya interaction

“…11 In spintronics, domain wall (DW) magnetic tunnel junction (MTJ) devices can be utilized to achieve multiple resistance states by stopping the DWs at different positions. 12 However, DW devices exhibit stochastic DW motion and it is very difficult to pin the DWs at certain positions in a controlled and repeatable manner. 13 In the past, researchers have studied the concept of artificial pinning sites to controllably and reproducibly pin the DWs at certain positions along the nanowire.…”