Tuning a binary ferromagnet into a multi-state synapse with spin-orbit torque induced plasticity
Yi Cao, Andrew Rushforth, Yu Sheng, Houzhi Zheng, and Kaiyou Wang

TL;DR
This paper demonstrates how a binary ferromagnet can be tuned into a multi-state synapse using spin-orbit torque without external magnetic fields, enabling neuromorphic computing functionalities.
Contribution
It introduces a novel method to achieve multi-state magnetic switching in a ferromagnetic structure via interlayer exchange coupling and spin-orbit torque, without external magnetic fields.
Findings
Achieved current-driven multi-state magnetization switching.
Demonstrated synaptic plasticity functionalities like EPSP, IPSP, and STDP.
Numerical simulations explained the broadening of magnetic reversal characteristics.
Abstract
Inspired by ion-dominated synaptic plasticity in human brain, artificial synapses for neuromorphic computing adopt charge-related quantities as their weights. Despite the existing charge derived synaptic emulations, schemes of controlling electron spins in ferromagnetic devices have also attracted considerable interest due to their advantages of low energy consumption, unlimited endurance, and favorable CMOS compatibility. However, a generally applicable method of tuning a binary ferromagnet into a multi-state memory with pure spin-dominated synaptic plasticity in the absence of an external magnetic field is still missing. Here, we show how synaptic plasticity of a perpendicular ferromagnetic FM1 layer can be obtained when it is interlayer-exchange-coupled by another in-plane ferromagnetic FM2 layer, where a magnetic-field-free current-driven multi-state magnetization switching of FM1…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
