P3BT Organic Memristor-Based Artificial Synapse for Neuromorphic Computing
Hongguang Zhang, Junxin Liu, Rui Shi, Wei Li, Xinran Li, Xiaoyu Hou, Yiheng Sun, Liqing Liu, Mingdong Yi, Wen Li, Yongtao LiAbstract
Organic memristors have potential applications in simulating biological synapses for neuromorphic computing. In this work, a solution-processed Al/P3BT/indium tin oxide (ITO) organic memristor was designed. The device exhibits stable resistive switching behavior and successfully emulates versatile synaptic plasticity, including excitatory postsynaptic current (EPSC), paired-pulse facilitation (PPF), spike-timing-dependent plasticity (STDP), the transition from short-term plasticity to long-term potentiation (STP-LTP), post-tetanic potentiation (PTP), and experiential learning. Fitting analyses of the I–V curves reveal that the resistive switching resulted from a combined contribution of direct tunneling, space-charge-limited conduction, Schottky emission, and Fowler–Nordheim tunneling. Moreover, the analog resistive switching behavior remains stable, even after exposure to ambient air for over 300 days. To exploit the nonlinear mappings through simple 3-bit pulse sequences, the constructed reservoir computing system achieves a recognition accuracy of 97.69% on the Modified National Institute of Standards and Technology (MNIST) handwritten digit classification task after only 20 training epochs, significantly outperforming conventionally trained network models. Furthermore, the system attains an accuracy of 87.02% on the Fashion-MNIST data set. This study provides valuable insights and perspectives on the emulation of artificial synaptic plasticity by using organic memristors and their application in neuromorphic computing.