Van der Waals Antiferroelectric CuCrP₂S₆-Based Artificial Synapse for High-Precision Neuromorphic Computation

Area of Science:

• Materials Science
• Nanotechnology
• Neuroscience

Background:

• 2D van der Waals heterostructures offer tunable properties for artificial synapses.
• Neuromorphic systems require efficient synaptic devices for advanced computing.

Purpose of the Study:

• To develop and characterize a novel 2D van der Waals antiferroelectric field-effect transistor (AFe-FET) for synaptic applications.
• To emulate neuroplasticity and evaluate the device's performance in neuromorphic computing tasks.

Main Methods:

• Fabrication of a SnS₂/h-BN/CuCrP₂S₆ AFe-FET.
• Investigation of synaptic weight modulation via charge trapping and ferroelectric polarization.
• Emulation of paired-pulse facilitation, short-term, and long-term plasticity.

Main Results:

• The AFe-FET successfully emulated neuroplasticity features.
• Achieved exceptional long-term potentiation/depression with low nonlinearity (1.1) and high dynamic range (10).
• Demonstrated 97.7% accuracy on MNIST and 94.7% in cognitive simulations.

Conclusions:

• The AFe-FET presents a new paradigm for high-fidelity synaptic devices.
• Offers a strategy for energy-efficient, biologically plausible neuromorphic computing.
• Highlights the potential of engineered 2D materials in next-generation AI.