WSe2/graphene heterojunction synaptic phototransistor with both electrically and optically tunable plasticity

Yilin Sun, Yuxuan Lin, Ahmad Zubair, Dan Xie*, Tomás Palacios*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

22 Citations (Scopus)

Abstract

The imitation of synaptic plasticity in artificial neuromorphic devices has been widely realized based on memristors, transistors and ion devices. This development of artificial synaptic devices is expected to open up a new era for neuromorphic computing. However, the complicated functions in biological synapse are dependent on the dynamic neural activities with modulated plasticity, which is still very difficult to emulate at the device level. Here, an artificial synaptic transistor based on WSe2/graphene van der Waals heterojunction is demonstrated with both electrically and optically modulated synaptic plasticity. By changing the polarity of applied V gs and V ds as well as superimposed gate voltage spikes, both excitatory and inhibitory synaptic plasticity can be realized in a single device. Moreover, due to the asymmetric optical response of WSe2/graphene heterojunction, optical modulation on hysteretic behaviors is achieved and also contributes to the inversion of synaptic plasticity. Hence, this work demonstrates a coordinated regulation on synaptic plasticity under electrical and optical stimuli and provides a multi-dimensional modulation strategy to understand the underlying mechanisms of complicated neuromorphic computing and further develop multifunctional neuromorphic optoelectronics.

Original languageEnglish
Article number035034
Journal2D Materials
Volume8
Issue number3
DOIs
Publication statusPublished - Jul 2021
Externally publishedYes

Keywords

  • WSe
  • grapheme
  • plasticity modulation
  • synaptic phototransistor

Fingerprint

Dive into the research topics of 'WSe2/graphene heterojunction synaptic phototransistor with both electrically and optically tunable plasticity'. Together they form a unique fingerprint.

Cite this