Substrate Orientation-Dependent Synaptic Plasticity and Visual Memory in Sol-Gel-Derived ZnO Optoelectronic Devices

Area of Science:

• Materials Science
• Neuroscience
• Electrical Engineering

Background:

• Optoelectronic synaptic devices mimic biological synapses for neuromorphic computing.
• ZnO-based devices offer promising properties for artificial intelligence hardware.
• Substrate selection is crucial for optimizing device performance.

Purpose of the Study:

• To investigate the influence of substrate orientation (c-plane vs. m-plane sapphire) on Al/ZnO/Al optoelectronic synaptic devices.
• To analyze the impact of crystallographic effects on synaptic behaviors and memory retention.
• To demonstrate the practical application of these devices in optoelectronic synaptic arrays.

Main Methods:

• Fabrication of Al/ZnO/Al synaptic devices using a sol-gel process on c-plane and m-plane sapphire substrates.
• Characterization of essential synaptic behaviors including excitatory postsynaptic current (EPSC) modulation, paired-pulse facilitation, and learning-forgetting dynamics.
• Construction and testing of 3x3-pixel optoelectronic synaptic arrays to demonstrate pattern encoding and retention.

Main Results:

• Devices exhibit key synaptic functionalities: EPSC modulation, paired-pulse facilitation, and Wickelgren's power law-described learning-forgetting.
• M-plane sapphire substrates yield devices with higher EPSCs, slower decay rates, and superior memory retention compared to c-plane.
• Crystallographic effects on m-plane substrates enhance carrier trapping and persistent photoconductivity, improving memory performance.
• Successful encoding, learning, and retention of an optical pattern within a 3x3-pixel array were demonstrated.

Conclusions:

• Substrate orientation is a critical factor in tailoring synaptic plasticity and memory retention in ZnO-based optoelectronic synapses.
• M-plane sapphire offers superior performance for optoelectronic synaptic devices, enhancing their suitability for neuromorphic computing.
• These findings pave the way for advanced ZnO-based optoelectronic synaptic arrays in in-sensor computing and artificial visual memory systems.