Proceedings of MATSUS Spring 2025 Conference (MATSUSSpring25)
Publication date: 16th December 2024
Neuromorphic computing systems, which mimic the neural architecture of the human brain, offer superior energy efficiency and processing speed compared to conventional von Neumann architectures, making them highly suitable for large-scale data processing [1]. A fundamental building block of these systems is the artificial synapse, which replicates synaptic plasticity—the dynamic modulation of synaptic strength in response to external stimuli. This adaptability is crucial for enabling neural networks to learn and retain information, supporting both short-term and long-term memory formation.
Organic memristors have emerged as promising candidates for flexible artificial synapses due to their mechanical compliance and capability to gradually modulate resistance states under electrical stimuli, effectively emulating synaptic plasticity [2]. Their functionality relies on the formation and dissolution of conductive filaments (CFs), allowing tunable synaptic weight modulation. However, practical implementation remains challenging, often necessitating intricate circuit designs and additional engineering considerations [3].
In this study, we introduce a small molecule-based organic memristor as a flexible artificial synapse, designed to precisely regulate CF growth pathways. This strategy enables selective emulation of both short-term and long-term plasticity, offering a versatile platform for implementing diverse neural network learning paradigms in flexible systems. The proposed approach is expected to enhance the learning and memory capabilities of artificial intelligence systems, contributing to the advancement of next-generation neuromorphic and wearable electronics.
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2024-00411764)