High-Speed and Low-Energy Resistive Switching with Two-Dimensional Cobalt Phosphorus Trisulfide for Efficient...

Area of Science:

• Materials Science
• Nanotechnology
• Computer Engineering

Background:

• Two-dimensional (2D) materials are promising for neuromorphic computing due to their electrical and mechanical properties.
• Implementing 2D memristors faces challenges in balancing low latency and low energy consumption.

Purpose of the Study:

• To demonstrate memristors based on 2D cobalt phosphorus trisulfide (CoPS3) for neuromorphic applications.
• To achieve high performance metrics including speed, energy efficiency, and reliability.

Main Methods:

• Fabrication of CoPS3 memristors.
• Electroforming process to create sulfur vacancies and conductive filaments.
• Characterization of switching speed, energy consumption, switching ratio, and voltage requirements.
• Evaluation of conductance modulation linearity and memory retention.
• Testing in artificial neural network models for handwritten digit recognition and image processing.

Main Results:

• CoPS3 memristors exhibit high switching speed (20 ns) and low switching energy (1.15 pJ).
• Achieved high switching ratio (>400) and low switching voltages (1.05 V set, -0.89 V reset).
• Demonstrated linear conductance modulation, long-term memory retention, and successful application in neural network models.
• Robust switching observed in solution-processed, large-scale CoPS3 films.

Conclusions:

• 2D CoPS3 materials show significant potential for energy-efficient neuromorphic computing circuits.
• The developed memristors offer a combination of rapid switching, low energy use, and scalability.
• Solution-processed CoPS3 films enable wafer-scale, low-temperature integration for advanced computing architectures.