High-efficiency polarization beam merging metagrating design via global initial solution-based topology optimization.

Area of Science:

• Electromagnetics and Optics
• Materials Science
• Computational Physics

Background:

• Metasurfaces offer precise electromagnetic control for applications like metagratings and metalenses.
• Initial structure selection critically impacts topology optimization outcomes for metasurface design.

Purpose of the Study:

• To propose and validate a global initial solution-based topology optimization (GISTO) method for efficient polarization beam merging metagratings.
• To enhance the design efficiency and performance of polarization beam merging metagratings through integrated optimization strategies.

Main Methods:

• Employed a 2D encoded non-dominated sorting genetic algorithm II for global exploration to find a high-quality initial structure.
• Utilized gradient-based topology optimization for local refinement of the initial structure.
• Integrated global and local optimization techniques for comprehensive design.

Main Results:

• Achieved significantly improved efficiency in polarization beam merging metagratings.
• Under symmetric incidence, efficiencies reached 93.6% (x-polarized) and 95.3% (y-polarized).
• Under asymmetric incidence, efficiencies reached 99.4% (x-polarized) and 95.4% (y-polarized).

Conclusions:

• The GISTO method effectively designs high-performance polarization beam merging metagratings.
• Integrating global and local optimization is a powerful strategy for advancing metasurface functionalities.
• The proposed method demonstrates superior efficiency for polarization beam merging applications.