Two-level optimizer for large-scale metasurfaces with strong near-field coupling

Area of Science:

• Optics and Photonics
• Materials Science
• Computational Physics

Background:

• Metasurfaces offer novel optical functionalities but designing them with high performance, especially for low aspect-ratio structures, faces fabrication-performance trade-offs.
• Accurate modeling of near-field coupling effects is crucial for optimizing metasurface performance but computationally intensive.
• Inverse-design approaches are powerful for optimizing complex structures like metasurfaces.

Purpose of the Study:

• To introduce a novel two-level window-based optimization architecture for the inverse-design of metasurfaces.
• To accurately model near-field coupling effects during the optimization process.
• To address the established fabrication-performance tradeoff for low aspect-ratio metasurfaces.

Main Methods:

• A two-level window-based optimization architecture with two nested iterative optimizers was developed.
• The method enables the inverse-design of metasurfaces with hundreds of adjustable parameters.
• Near-field coupling effects were accurately modeled throughout the optimization process.

Main Results:

• The architecture effectively optimized a 300 unit-cell metalens with a 108 µm aperture diameter.
• A 73.2% increase in focusing intensity was achieved compared to conventional designs.
• The method successfully addressed the fabrication-performance tradeoff for low aspect-ratio metasurfaces.

Conclusions:

• The proposed two-level window-based optimization architecture is effective for inverse-design of complex metasurfaces.
• This approach significantly enhances focusing intensity while managing near-field coupling.
• The method provides a pathway to overcome limitations in metasurface fabrication and performance for specific applications.