PEDOT-ZnO Nanoparticle Hybrid Film-Based Memristors for Synapse Emulation in Neuromorphic Computing Applications

Synapses and neurons in artificial intelligence are acknowledged as pivotal elements in constructing neuromorphic computing systems. Specifically, organic material-based memristors are widely recognized due to their advantages in terms of transparency, flexibility, cost-effectiveness, environment friendliness, and biocompatible properties. In this paper, we fabricated organic functional layer-based memristors and synapses by spin-coating poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate) (PEDOT:PSS) onto an indium tin oxide (ITO) substrate following a magnetron-sputtered ITO as the top electrode. Moreover, zinc oxide nanoparticles (ZnO NPs) were employed in the functional layer as charge trapping elements for optimizing the performance of PEDOT-based memory devices. With control of the concentrations of ZnO NPs, the devices of ITO/PEDOT:PSS(ZnO NPs)/ITO exhibited promising resistive switching performance and synaptic functionalities. In the device of ITO/PEDOT:PSS(3% ZnO NPs)/ITO, it shows gradual switching characteristics and could effectively mimic synapse functionalities. In the device ITO/PEDOT:PSS(5% ZnO NPs)/ITO, it shows a forming-free character and excellent resistive change performance.