Journal Home > Volume 16 , Issue 1

Photoelectric synaptic devices have been considered as one of the key components in artificial neuromorphic systems due to their excellent capability to emulate the functions of visual neurons, such as light perception and image processing. Herein, we demonstrate an optically-stimulated artificial synapse with a clear photoresponse from ultraviolet to visible light, which is established on a novel heterostructure consisting of monocrystalline Cs2AgBiBr6 perovskite and indium–gallium–zinc oxide (IGZO) thin film. As compared with pure IGZO, the heterostructure significantly enhances the photoresponse and corresponding synaptic plasticity of the devices, which originate from the superior visible absorption of single-crystal Cs2AgBiBr6 and effective interfacial charge transfer from Cs2AgBiBr6 to IGZO. A variety of synaptic behaviors are realized on the fabricated thin-film transistors, including excitatory postsynaptic current, paired pulse facilitation, short-term, and long-term plasticity. Furthermore, an artificial neural network is simulated based on the photonic potentiation and electrical depression effects of synaptic devices, and an accuracy rate up to 83.8% ± 1.2% for pattern recognition is achieved. This finding promises a simple and efficient way to construct photoelectric synaptic devices with tunable spectrum for future neuromorphic applications.

File
12274_2022_4806_MOESM1_ESM.pdf (3 MB)
Publication history
Copyright
Acknowledgements

Publication history

Received: 20 April 2022
Revised: 20 July 2022
Accepted: 25 July 2022
Published: 17 September 2022
Issue date: January 2023

Copyright

© Tsinghua University Press 2022

Acknowledgements

Acknowledgements

Shi Y. M. and Han C. acknowledge the support from the National Natural Science Foundation of China (Nos. 61874074 and 62004128), the Fundamental Research Foundation of Shenzhen (Nos. JCYJ20170817101100705 and JCYJ20190808152607389) and the Key Project of Department of Education of Guangdong Province (No. 2016KZDXM008). Li H. N. acknowledges the support from the Natural Science Foundation of SZU (No. 2017011) and the Science and Technology Project of Shenzhen (No. JCYJ20170817100111548). This research is supported by Singapore Ministry of Education under its AcRF Tier 2 (No. MOE-T2EP50220-0001), the Shenzhen Peacock Plan (No. KQTD2016053112042971), and the postgraduate innovation development fund project of Shenzhen University (No. 315-0000470527).

Return