Bidirectional All-Optical Synapses Based on a 2D Bi2O2Se/Graphene Hybrid Structure for Multifunctional Optoelectronics

Chia Ming Yang, Tsung Cheng Chen, Dharmendra Verma, Lain Jong Li, Bo Liu, Wen Hao Chang, Chao Sung Lai*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Neuromorphic computing has been extensively studied to mimic the brain functions of perception, learning, and memory because it may overcome the von Neumann bottleneck. Here, with the light-induced bidirectional photoresponse of the proposed Bi2O2Se/graphene hybrid structure, its potential use in next-generation neuromorphic hardware is examined with three distinct optoelectronic applications. First, a photodetector based on a Bi2O2Se/graphene hybrid structure presents positive and negative photoresponsibility of 88 and −110 A W−1 achieved by the excitation of visible wavelength and ultraviolet wavelength light at intensities of 1.2 and 0.3 mW cm−2, respectively. Second, this unique photoresponse contributes to the realization of all optically stimulated long-term potentiation or long-term depression to mimic synaptic short-term plasticity and long-term plasticity, which are attributed to the combined effect of photoconductivity, bolometric, and photoinduced desorption. Third, the devices are applied to perform digital logic functions, such as “AND” and “OR,” using full light modulation. The proposed Bi2O2Se/graphene-based optoelectronic device represents an innovative and efficient building block for the development of future multifunctional artificial neuromorphic systems.

Original languageEnglish
JournalAdvanced Functional Materials
DOIs
StateAccepted/In press - 2020

Keywords

  • BiOSe
  • graphene
  • negative photoresponse
  • neuromorphics
  • optoelectronics

Fingerprint Dive into the research topics of 'Bidirectional All-Optical Synapses Based on a 2D Bi<sub>2</sub>O<sub>2</sub>Se/Graphene Hybrid Structure for Multifunctional Optoelectronics'. Together they form a unique fingerprint.

Cite this