Performance Study of Bulk Heterojunction Organic Synaptic Transistors Based on Polymer Blends

Neuromorphic computing, inspired by the intricate neural networks of the human brain, seeks to transcend the limitations of conventional computing by integrating information processing and storage. Central to this paradigm shift are artificial synaptic devices, the cornerstone of neuromorphic systems, which excel in energy efficiency by facilitating low-power signal processing. This attribute has garnered substantial interest within the artificial intelligence domain. A pivotal challenge in neuromorphic computing is the precise modulation of synaptic performance through innovative device architectures. This is vital for realizing neural computations with high precision and fidelity. Among the various strategies explored, bulk heterojunctions have garnered significant attention due to their distinctive structural and functional attributes, which offer precision and potential in fine-tuning synaptic behavior. In this work, a polymer-co-mixed bulk heterojunction transistor is proposed, which has been fabricated by merely combining a p-type semiconductor, which has an excess of positive charge carriers, with an n-type semiconductor, which has an excess of negative charge carriers, through a simple mixing process without the use of an additional capture layer, resulting in the fabrication of a bulk heterojunction organic synaptic transistor (BHJ-OST) with signalling and self-learning properties and a large storage window of more than 60V. In addition, the device successfully mimics biological synaptic properties, including excitatory postsynaptic current (EPSC), inhibitory postsynaptic current (IPSC), and long-range plasticity (LTP).