Voltage control of domain walls in magnetic nanowires for energy-efficient neuromorphic devices

Abstract

An energy-efficient voltage controlled domain wall device for implementing an artificial neuron and synapse is analyzed using micromagnetic modeling in the presence of room temperature thermal noise. By controlling the domain wall motion utilizing spin transfer or spin orbit torques in association with voltage generated strain control of perpendicular magnetic anisotropy in the presence of Dzyaloshinskii-Moriya interaction, different positions of the domain wall are realized in the free layer of a magnetic tunnel junction to program different synaptic weights. The feasibility of scaling of such devices is assessed in the presence of thermal perturbations that compromise controllability. Additionally, an artificial neuron can be realized by combining this DW device with a CMOS buffer. This provides a possible pathway to realize energy efficient voltage controlled nanomagnetic deep neural networks that can learn in real time.

Emerging Artificial Neuron Devices for Probabilistic Computing

“…This type of artificial synapse can be naturally coupled to either CMOS neurons or other artificial neurons to implement the firing behavior. Furthermore, other simulation approaches have been utilized to control the wall motion (Hassan et al, 2018;Azam et al, 2020). Although manipulating the nanosized skyrmion can yield multilevel states in ferromagnets (Azam et al, 2018;Chen et al, 2018;Liang et al, 2020), artificial synapses or neural components based on skyrmion have not been reported thus far.…”

Prospect of Spin-Orbitronic Devices and Their Applications

“…Although spin-orbitronic devices show better performance indexes over other memristor candidates, however, the characteristic binary resistance nature of generic MTJ makes it difficult to proceed implementing as a multilevel synapse rather than a stochastic binary synapse ( Grollier et al., 2020 ). One representative solution is to utilize the current-induced domain wall motion within the FM free layer ( Sengupta et al., 2015a ; Lequeux et al., 2016 ; Yue et al., 2019 ; Yang et al., 2019c ; Siddiqui et al., 2019 ; Azam et al., 2020 ; Zhang et al., 2019b ), for instance, by tuning the pinning potential of FM domain wall motions, SOT-induced multilevel magnetization switching as well as the typical synaptic functionality of spike-timing dependent plasticity (STDP) have been experimentally demonstrated ( Cao et al., 2019 ). Other strategies include the fine-magnetic domain switching in antiferromagnetic (AFM) ( Wadley et al., 2016 ; Olejník et al., 2017 ; Shi et al., 2020 ) or AFM/FM ( Liu et al., 2020b ; Zhou et al., 2020 ; Yun et al., 2020 ) heterostructures where multiple ∼100 nm-sized binary FM domains fixed by the polycrystalline AFM could reverse independently under the applying current ( Fukami et al., 2016 ; Kurenkov et al., 2017 ; Borders et al., 2016 ) and the SOT-induced skyrmion (a topological magnetic state) motions where the number of skyrmions within the signal reading area is proposed to represent the analog synaptic weight ( Song et al., 2020a ).…”

Voltage-Controlled Energy-Efficient Domain Wall Synapses With Stochastic Distribution of Quantized Weights in the Presence of Thermal Noise and Edge Roughness

“…Recently, nanomagnet based synaptic devices has shown potential to be energy efficient compared to PCRAM and RRAM [9,10,11]. Among nanomagnet based neuromorphic devices, domain wall (DW) based MTJs are one of the most promising.…”

“…Recently, strain-mediated control of DW has been reported [28,29]. Strain gradient in conjunction with SOT or STT [10] has also been proposed to control DW position to implement energy efficient synaptic devices that can be programmed in real time.…”