Floating-gate memristor based on a MoS₂/h-BN/AuNPs mixed-dimensional heterostructure

Area of Science:

• Materials Science
• Nanotechnology
• Electronics Engineering

Background:

• Memristors are crucial for neuromorphic computing, mimicking brain function for faster computation.
• Traditional memristors (e.g., MoO3, HfO2) face challenges with low switching ratios and poor stability.
• Developing advanced memristors is key to overcoming current limitations in electronic circuits.

Purpose of the Study:

• To develop a novel floating-gate memristor utilizing a mixed-dimensional heterostructure.
• To evaluate the device's performance for neuromorphic computing and multibit storage applications.
• To investigate the underlying mechanisms and assess the durability and stability of the new memristor design.

Main Methods:

• Fabrication of a heterostructure memristor with 2D MoS2, 0D Au nanoparticles (AuNPs), and an h-BN insulating layer.
• Characterization of device operation under back-gate voltage modulation, analyzing resistance states (HRS, LRS).
• Testing durability (>3000 cycles) and stability (retaining states >10^4 s), and evaluating responses to electrical/optical pulses.

Main Results:

• The MoS2/h-BN/AuNPs memristor reliably switches between high-resistance state (HRS) and low-resistance state (LRS).
• The device exhibits multiple stable LRS states, indicating potential for multibit storage.
• Achieved an on/off ratio >10^4 with excellent durability and stability, retaining states >10^4 s over 3000 cycles.

Conclusions:

• The developed floating-gate memristor demonstrates significant potential for advanced neuromorphic computing and multibit data storage.
• Electron quantum tunneling between AuNPs and the MoS2 channel governs the device's modulation.
• The memristor's robust performance, including high durability and stability, marks a substantial advancement over traditional devices.