Model of Organic Ferroelectric Transistors for Neural Networks

FeFETs are promising for memory computing in neural networks. Organic ferroelectric materials offer benefits like affordability, scalability, flexibility, and low-temperature processing over their inorganic counterparts. This work introduces a sophisticated physics-based OFeFET model that captures the relationship between capacitive layer thickness and coercive electric field in organic ferroelectric capacitors. The model, validated against a 300 nm P(VDF-TrFE) OFeFET, was integrated into Neurosim to evaluate image recognition performance. Achieving up to 94% accuracy, this work bolsters the potential of organic memory devices in computational applications.