Photopatternable polypyrrole electrode-based organic optical synaptic transistor for reservoir computing

Abstract

Inspired by the efficient information processing of biological neural systems, optoelectronic synaptic devices integrating sensing and computation have emerged as a cornerstone for next-generation neuromorphic computing. However, achieving high-performance organic synaptic transistors remains challenging due to the limitations of conventional devices in terms of patterning precision. Here, we develop an organic tunable-plasticity transistor (OTPT) based on photolithographic polypyrrole (PPy) electrode patterns. The PPy electrodes exhibit exceptional adhesion, wafer-scale uniformity, and high resolution. Upon depositing pentacene as the semiconductor layer on the octadecyltrichlorosilane-modified substrate, the resulting transistor devices emulate essential synaptic behaviors. Furthermore, we implement a physical reservoir computing (RC) system based on the OTPT to tackle complex multi-modal tasks. The system demonstrates high recognition accuracy in static digital recognition and dynamic sequence classification and successfully identifies bus motion direction. This work demonstrates that the high-precision integration of PPy electrodes provides a robust hardware foundation for the RC system, facilitating the development of energy-efficient neuromorphic electronics for edge intelligence.