Atomic Layer Deposited Hf_0.5Zr_0.5O₂-based Flexible Memristor with Short/Long-Term Synaptic Plasticity

Area of Science:

• Materials Science
• Neuroscience
• Computer Engineering

Background:

• Neuromorphic computing aims to mimic the human brain's structure and function.
• Traditional computing faces limitations (von Neumann bottleneck) in handling complex tasks.
• Artificial synapses are crucial components for building brain-inspired computing systems.

Purpose of the Study:

• To develop a flexible artificial synapse for neuromorphic computing.
• To investigate the synaptic behaviors and memory functions of the proposed device.
• To demonstrate the potential of memristors in next-generation computing architectures.

Main Methods:

• Fabrication of a flexible electrical synapse using low-temperature atomic layer deposition.
• Characterization of bipolar resistive switching behavior.
• Emulation of synaptic plasticity (short-term and long-term) and forgetting using pre-synaptic spikes.

Main Results:

• The device exhibited gradual conductance modulation via ion conductive filament formation and rupture.
• Demonstrated successful emulation of short-term plasticity, long-term plasticity, and forgetting behaviors.
• Integrated memory and learning functionalities into a single flexible memristor.

Conclusions:

• The developed flexible electrical synapse shows promise for neuromorphic computing applications.
• This work contributes to overcoming the limitations of conventional computing systems.
• The memristor-based approach offers a pathway towards efficient artificial intelligence hardware.