An organic optoelectronic synaptic transistor for image preprocessing

Abstract

Optoelectronic synapses, with their ability to sense and memorize optical signals, are pivotal in the advancement of bionic visual systems. However, current devices often suffer from suboptimal paired-pulse facilitation (PPF), hindering their application in real-time visual information encoding. In this study, we present an easily fabricated single-component p-type organic optoelectronic synaptic transistor that achieves a remarkable PPF index of approximately 249%. This is accomplished by incorporating an organic polymer PM6 into an electrolyte gated field-effect transistor frame. Significantly, the photogenerated carrier dynamics in PM6 can be modulated and elucidated through interactions with ionic liquids, thereby enhancing the device performance. This device mimics biological synaptic plasticity and demonstrates significant potential for image preprocessing, achieving a classification accuracy of 95.6% after denoising. Additionally, by leveraging the spatiotemporal decoding capabilities of the device, a physical node Reservoir Computing system was constructed. This work provides a viable strategy for advancing artificial vision systems that utilizes optical synaptic arrays.