Multilevel resistive switching memristor based on silk fibroin/graphene oxide with image reconstruction functionality

Shubin Liu,Yu Cheng,Fang Han,Suna Fan,Yaopeng Zhang
DOI: https://doi.org/10.1016/j.cej.2023.144678
IF: 15.1
2023-09-01
Chemical Engineering Journal
Abstract:Multilevel storage memristors have great potential for use in neuromorphic computing, high-density storage, and the creation of artificial synapses. In this work, we report a multilevel storage memristor based on graphene oxide (GO)/silk fibroin (SF)/GO memristive layer structure. This memristor incorporates binary and ternary switching behaviors in a single device. It was found that the transition of the behavior of the memristor can be transformed between the two resistive switching modes by regulating the compliance current (Icc ) applied to the device. Both switching behaviors are stable, repeatable, and nonvolatile. Furthermore, the device shows great potential in simulating synaptic plasticity and is applicable to use in artificial neural networks for digital image recognition as well as image compression and reconstruction. The highest accuracy of recognition of handwritten digital images based on the ternary neural network built by the device is as high as 92.3%. This work highlights GO/SF/GO memristors as promising devices for improving the storage density of memory cells and simplifying the structure of the memristor-based storage system.
engineering, chemical, environmental
What problem does this paper attempt to address?