Tunable synaptic plasticity in MoS2 neuromorphic transistors using Li+ incorporated chitosan electrolytes

Abstract

Neuromorphic computing hardware based on synaptic transistors can outperform traditional devices in energy-efficient computing and parallel data processing. Electric double-layer transistors with solid polymer electrolytes are particularly promising due to their simple fabrication, structural stability, and tunable properties. However, effective methods to tune synaptic response times across various applications using a single device structure remain limited. In this study, we explore the incorporation of lithium salts into chitosan-based electrolytes to enhance the performance of MoS₂ synaptic transistors. The addition of Li⁺ improves ionic conductivity within the electrolyte through a plasticizing effect and an increased ion concentration. Devices with Li⁺-incorporated electrolytes exhibit significantly higher double-layer capacitance, reduced subthreshold swing, and improved operation frequencies compared to those without Li⁺-incorporation. They also exhibit stronger synaptic responses with faster dynamics, including paired-pulse facilitation reaching ∼500%. Importantly, this enhancement is tunable by adjusting Li⁺ concentration, allowing repeated electrolyte replacement on the same device. Furthermore, using a dual-gate structure further improves the flexibility and controllability of the synaptic transistors. This strategy for regulating synaptic plasticity enables diverse neuromorphic functionalities within a single device structure and helps mitigate non-uniform performance in arrays.