Tunable Negative and Positive Photoconductance in Van Der Waals Heterostructure for Image Preprocessing

Area of Science:

• Materials Science
• Optoelectronics
• Neuroscience

Background:

• Visual information processing relies on retinal preprocessing for enhanced transmission quality and efficiency.
• Artificial retina systems offer a promising approach for efficient image processing.
• 2D materials offer unique electronic and optical properties for advanced device applications.

Purpose of the Study:

• To propose and investigate a novel graphene/InSe/h-BN heterogeneous structure for artificial vision applications.
• To explore the negative and positive photoconductance (NPC and PPC) effects in the proposed 2D material system.
• To simulate and validate the retinal preprocessing function using the developed artificial retina system.

Main Methods:

• Fabrication of a graphene/InSe/h-BN heterogeneous structure.
• Investigation of NPC and PPC effects under varying laser power.
• Development of a modified theoretical model to explain the observed photoconductance phenomena.
• Characterization of the field-effect transistor (FET) device performance and stability.
• Simulation of retinal preprocessing functions using the device.

Main Results:

• The graphene/InSe/h-BN structure exhibited distinct NPC and PPC effects.
• A theoretical model was established to elucidate the underlying physical mechanisms.
• The FET device demonstrated excellent photoelectric performance (R_NPC = 1.1×10⁴ AW^-1, R_PPC = 13 AW^-1) and stability.
• Successful simulation of retinal preprocessing functions was achieved.
• Pulse signal input enhanced device responsivity by 167%.

Conclusions:

• The proposed 2D heterogeneous structure provides a new design for artificial vision systems.
• The observed NPC and PPC effects are crucial for mimicking retinal preprocessing.
• The developed artificial retina system enhances visual signal transmission quality and efficiency.
• This work lays the foundation for next-generation optoelectronic devices integrated with artificial vision capabilities.