Related Experiment Video
Updated: Jun 27, 2025
13:02
Fabrication of 1-D Photonic Crystal Cavity on a Nanofiber Using Femtosecond Laser-induced Ablation
Published on: February 25, 2017
9.7K
Inverse Design of Photonic Surfaces via High throughput Femtosecond Laser Processing and Tandem Neural Networks
Minok Park1, Luka Grbčić2, Parham Motameni3
1Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, CA, 94720, USA.
Advanced Science (Weinheim, Baden-Wurttemberg, Germany)
|April 30, 2024
Summary
This study combines laser processing with AI to create novel photonic surfaces. The method efficiently designs textured substrates with specific optical properties, advancing energy harvesting technology.
Area of Science:
- Materials Science
- Optics
- Artificial Intelligence
Background:
- Designing photonic surfaces with tailored optical properties is crucial for energy harvesting.
- Traditional methods for surface design are often time-consuming and lack efficiency.
Purpose of the Study:
- To develop an inverse design method for photonic surfaces by integrating femtosecond laser processing and neural networks.
- To accelerate the discovery and fabrication of microtextured surfaces with specific spectral emissivities.
Main Methods:
- Utilized femtosecond laser processing to create 35,280 unique microtextured surfaces on stainless steel.
- Developed high-throughput fabrication and characterization platforms to generate a comprehensive dataset.
- Employed tandem neural networks for inverse design, linking laser parameters to spectral emissivity.
Main Results:
- The neural network model accurately predicted spectral emissivities (average RMSE < 2.5%) for novel designs.
- The inverse design approach identified optimal laser parameters within a significantly reduced parameter space (25x smaller).
- Successfully validated the model on a thermophotovoltaic emitter design.
Conclusions:
- The synergistic approach of laser-matter interaction and neural networks accelerates photonic surface discovery.
- This method offers a pathway for advancing energy harvesting technologies through efficient material design.
- Demonstrated the potential of AI-driven inverse design in materials science and optics.
