Solution Processable Mos2 Based Memristive Synapse for Brain Inspired Computing

Hanrui Li, Dayanand Kumar, Nazek Elatab

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

As a novel 2D material, MoS 2 has shown excellent electrical properties and resistive switching characteristics to work as a switching layer for non-volatile memory. In this work, we drop cast the MoS 2 solution to prepare the thin film and deposit an interfacial layer of Al 2 O 3 . We demonstrate the proposed memristive device with Cu/Al 2 O 3 /MoS 2 /Pt structure to work as an artificial synapse. The device shows a steady resistive switching behavior with the SET and RESET voltages of 1.3 V and -0.5 V, respectively. We further demonstrate the synapse behavior via a Hopfield Neural Network (HNN) and achieve image recognition and reconstruction with a high accuracy of 96% after 15 training epochs.
Original languageEnglish (US)
Title of host publication2023 IEEE 23rd International Conference on Nanotechnology (NANO)
PublisherIEEE
DOIs
StatePublished - Jul 2 2023

Fingerprint

Dive into the research topics of 'Solution Processable Mos2 Based Memristive Synapse for Brain Inspired Computing'. Together they form a unique fingerprint.

Cite this