MoS₂ Memristors Exhibiting Variable Switching Characteristics toward Biorealistic Synaptic Emulation

Area of Science:

• Materials Science
• Nanotechnology
• Neuroscience

Background:

• Memristors based on 2D layered materials offer potential for energy-efficient artificial neural networks by mimicking biological neuronal interactions.
• Systematic analysis of memristive switching mechanisms is crucial for developing reliable 2D-material-based memristors for neuromorphic computing.

Purpose of the Study:

• To investigate the switching characteristics of few-layer molybdenum disulfide (MoS2) memristors fabricated by mechanical printing.
• To understand the underlying physical mechanisms governing memristive behavior in MoS2 devices.
• To explore the potential for ionic coupling between MoS2 memristors for building neuromorphic systems.

Main Methods:

• Fabrication of few-layer MoS2 memristors using mechanical printing.
• DC and pulse programming to analyze memristive switching characteristics.
• Kelvin probe force microscopy, Auger electron spectroscopy, and electronic characterization to support findings.
• Fabrication of a testing device with two adjacent MoS2 memristors to demonstrate ionic coupling.

Main Results:

• Observed two types of DC-programmed switching: rectification-mediated and conductance-mediated, linked to Schottky barriers and vacancy redistribution.
• Demonstrated conversion of analog switching to quasi-binary switching via electrical stress, attributed to field-induced vacancy agglomeration.
• Successfully fabricated and tested a device showing ionic coupling between two MoS2 memristors.

Conclusions:

• The study provides critical insights into the device physics of 2D-material-based memristors, specifically MoS2.
• The observed ionic coupling mechanism is a promising pathway for creating neuromorphic computing systems that emulate biological neural networks.
• This work lays a foundation for advancing biorealistic neuromorphic computing using 2D layered materials.