An Opto-Iontronic Cholesteric Liquid Crystalline Retina for Multimodal Circularly Polarized Neuromorphic Vision.
basic_science · Level V
Where this comes from
- Record sourced from PubMed, PMID 40025907.
- Also identified by DOI 10.1002/adma.202419747.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
Circularly polarized light (CPL) is fundamental to phase-controlled imaging, quantum optics, and optical computing. Conventional CPL detection, relying on polarizers and quarter-wave plates, complicates device design and reduces sensitivity. Among emerging CPL detectors, organic field-effect transistors (OFET) with helical organic semiconductors are highly promising due to their compact structures but suffer tedious synthesis, low dissymmetric factors (g<sub>ph</sub> < 0.1), and high operating voltages (> 50 V). To address these issues, an opto-iontronic cholesteric liquid crystalline (i-CLC) film is developed that is both electrically and photonically active, serving as the dielectric in phototransistors. The well-defined cholesteric structure and broadly tunable pitches of the i-CLC film enable it to detect CPL with an excellent "handedness" selectivity across a broad spectrum. Moreover, its ionic nature provides a high capacitance (up to 580 nF cm<sup>-</sup> <sup>2</sup> @20 Hz). The resulting flexible CPL detectors achieve an unprecedentedly high dissymmetry factor (g<sub>ph</sub> = 1.33) at low operating voltages (< 5 V), showcasing significant potential in optical communication and data encryption. Furthermore, leveraging high g<sub>ph</sub>, they can perform in-sensor computing for highly accurate semantic segmentation using fused multimodal visual inputs (e.g., circularly polarized and ordinary light), achieving an accuracy of 75.73% and a mean intersection over the union of 0.3982, surpassing the performance of non-CPL photodetectors. Additionally, it optimizes power consumption by a factor of 10<sup>2</sup> compared to most conventional visual processing systems, offering a groundbreaking hardware solution for high-performance neuromorphic CPL vision.