Application and challenges of organic electrochemical transistors in neuromorphic computing: bionic synapse and multi-mode integration

Abstract

Organic electrochemical transistors (OECTs) have emerged as promising candidates for neuromorphic computing due to their mixed ion–electron conduction mechanism and excellent biocompatibility, enabling bio-inspired synaptic devices. Recently, OECTs have made advances in replicating synaptic plasticity, highlighting their low operating voltages, high transconductance, and multi-level conductance modulation. In this review, we first introduce the structural characteristics of OECTs and their coupling mechanisms with mixed ion–electron conductors (MIECs). Next, we elaborate developments in enhancement-mode OECTs for biomimetic synapses, including stretchable devices that mimic short- and long-term plasticity, light-responsive synapses regulated by optical pulses, and fully printed complementary circuits for real-time biosensing. Third, we discuss depletion-mode OECTs, focusing on their dynamic inhibitory mechanisms and emerging applications in synaptic inhibition. Fourth, we discuss three representative models for the device-to-network mapping strategies and performance advantages of OECTs in experimentally validated neural network architectures. Finally, a conclusion is drawn, addressing the challenges and prospects faced by OECTs. Nowadays, OECT-based synaptic research remains limited, requiring further efforts in the electrochemical stability of materials, modeling high-frequency responses, and addressing crosstalk in high-density arrays. These efforts could drive OECTs from single-point biomimicry to complex neural networks.