Neuromorphic Performance of Organic FeFETs Using Novel Ferroelectric Compounds: Towards In-Memory Computing

Neuromorphic computing architectures that mimic brain-like information processing offer significant advantages for energy-efficient artificial intelligence and in-memory computing. This work investigates the neuromorphic performance of organic FEFETs fabricated using novel solution-processable ferroelectric materials based on metals. Electrical characterization demonstrates stable ferroelectric switching with well-defined memory windows. Neuromorphic measurements reveal long-term potentiation/depression (LTP-LTD), spike-timing dependent plasticity (STDP), and paired-pulse facilitation/depression (PPF-PPD), confirming synaptic emulation capabilities. Memory retention and endurance testing show device longevity, while performing suitable handwritten pattern recognition. These insights will directly contribute to advancing neuromorphic device technology and accelerating the development of energy-efficient in-memory computing systems that can overcome the limitations of conventional von Neumann architectures.