Optoelectronic artificial synapses based on β-Ga2O3 films by RF magnetron sputtering

Optoelectronic neuromorphic architectures have promising potentials for energy-efficient computing of big data, which are urgently needed to satisfy the demands on the accurate perception and understanding for artificial intelligence techniques. In this work, optoelectronic artificial synapses based on a Pt/beta-Ga2O3/Pt structure were fabricated on MgO (111) substrates by RF magnetron sputtering. The optoelectronic plasticity was investigated in detail. It was found that the device can simulate the synaptic short-term memory and long-term memory, the transition from short-term memory to long-term memory, paired-pulse facilitation and learning-experience behavior. The device can detect and simulate synaptic behavior at different wavelengths of light, including visible to ultraviolet light. The optoelectronic artificial synapses realized the integration of detection and storage in one device, and the terse structure benefits for processing a large number of data with fast computational speed due to low crosstalk and low power consumption. Therefore, these optoelectronic neuromorphic architectures have great potentials and advantages to be applied in many fields, such as artificial intelligence, intelligent robots, and bioelectronics. Meanwhile, it will attract more attention to the study of optoelectronic synapses.

Automated summary

Optoelectronic artificial synapses based on Pt/beta-Ga2O3/Pt structure were fabricated and investigated for their optoelectronic plasticity, demonstrating capabilities in simulating synaptic memory states, learning behavior, and response to different light wavelengths. The compact structure with low power consumption and crosstalk allows for efficient processing of large data sets.