Unraveling Stochastic Dynamics and Switching Mechanism in Ag Network-Based Neuromorphic Device by Impedance Spectroscopy.
Journal:
Small (Weinheim an der Bergstrasse, Germany)
Published Date:
Jul 7, 2025
Abstract
Neuromorphic devices are leading advancements in brain-inspired computing. The present study employs impedance spectroscopy on an in-plane volatile neuromorphic device with self-formed silver (Ag) structures to understand its conduction mechanism. Ag islands act as synaptic junctions, while smaller Ag nanoparticles serve as signal transmission channels, mimicking a neural network. These devices hold promise for dynamic neural networks and reservoir computing. Electrical stimulation emulates synaptic functionalities, and impedance analysis confirms structural similarity via an equivalent RC circuit model. With device resistance decreasing by six orders of magnitude during the high resistance (HRS) to low resistance (LRS) transition, and the capacitance remaining in the picofarad (pF) range, suggests a metallic filamentary mechanism. AC field transitions, crucial for real-world applications, remain underexplored, and the present impedance-time analysis reveals sporadic conduction paths under a DC voltage, highlighting the role of the electric field. Impedance spectra at varying relative humidities (RH) reveal significant diffusion contributions at high RH, further analyzed via distribution of relaxation times (DRT) for electrochemical insights. X-ray photoelectron spectroscopy (XPS) supports the proposed mechanism. A successful equivalence with the biological synapse is established, resulting in a significant advancement in enhancing the understanding of the operational mechanism involved in neuromorphic devices.
Authors
Keywords
No keywords available for this article.
