Low energy consumption and stable perovskite-based photonic synapse with two-terminal lateral configuration

Fei Liu, Xiangyang Zhang, Agnes Valencia, Weilu Li, Xiangkun Bo, Walid A. Daoud*

*Corresponding author for this work

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

2 Citations (Scopus)

Abstract

Perovskites have unique photoelectric characteristics making them promising for use in artificial visual systems, which face the challenges of short-circuiting, high-energy consumption, and lack of in-depth knowledge on working mechanisms and stability. In this study, a novel two-terminal lateral configuration synapse based on perovskite/SnO2 heterojunction is introduced. This synapse exhibits typical synaptic characteristics with ultra-low energy consumption (0.1665 fJ), wide substrate selection, sensitivity to a broad range of wavelengths, and long-term operational stability. Additionally, the synapse potential application is demonstrated by emulating the classical conditioning and learning experience behavior, and neural network simulation. Further, the working mechanism based on photogeneration and trapping of charge carriers is revealed by analyzing the dynamic processes of photo-generated charge carriers, thus establishing a foundation for developing a simple yet efficient photonic synapse for artificial neuron devices. © 2025 Elsevier Ltd.
Original languageEnglish
Article number111145
JournalNano Energy
Volume142
Issue numberPart A
Online published14 May 2025
DOIs
Publication statusPublished - Sept 2025

Funding

This work was supported by City University of Hong Kong (Grant no. 7020038).

Research Keywords

  • Perovskite-based photonic synapse
  • Two-terminal lateral structure
  • Electron trapping
  • Low energy consumption
  • Artificial neural network

Fingerprint

Dive into the research topics of 'Low energy consumption and stable perovskite-based photonic synapse with two-terminal lateral configuration'. Together they form a unique fingerprint.

Cite this