Chirality, essential for distinguishing molecular enantiomers, impacts various scientific fields. Circular dichroism (CD) spectroscopy, which measures differential absorption of circularly polarized light, is limited by weak chiral-optical interactions, yielding minimal signals. Superchiral metasurfaces, enhancing optical chirality, offer improved CD signal and interaction with chiral molecules, necessitating advanced development for accurate analysis. This study presents a novel two-stage design methodology for optimizing superchiral metasurfaces, employing a neural network-based optimizer and adjoint topology optimization for enhanced optical chirality density. The results provide a platform for sensitive CD spectroscopy, facilitating fast optimization of metasurface unit cells for better chiral molecule analysis.
|