Enterprise AI Analysis

Polarization-sensitive neuromorphic vision sensing enabled by pristine black arsenic-phosphorus

This research introduces a groundbreaking polarization-sensitive neuromorphic phototransistor based on two-dimensional black arsenic-phosphorus (b-As0.2P0.8) nanosheets. Leveraging its intrinsic anisotropy, the device achieves exceptional optoelectronic performance with a peak responsivity of 2.88 A W⁻¹, a polarization ratio of 4.7, and a 40 dB dynamic range in the near-infrared communication band. Critically, it simulates synaptic behaviors—analogous to human neural responses to visual stimuli—with paired-pulse facilitation values up to 201% and gate-tunable short-term/long-term plasticity. This enables a hybrid optical-electronic neural network (HOENN) to achieve over 90% classification accuracy on Fashion-MNIST and 71.38% reconstruction accuracy on the Yale Face Database. The device also reconstructs hidden targets with high fidelity using Stokes parameter extraction and degree of linear polarization mapping, revealing intricate polarization features invisible to conventional imaging. This work lays a foundational platform for high-performance neuromorphic vision systems, integrating polarization imaging, computation, and communication functionalities for scalable brain-inspired optoelectronic technologies.