Related Experiment Video
Updated: Jul 7, 2026
14:58
Optical Scatter Microscopy Based on Two-Dimensional Gabor Filters
Published on: June 2, 2010
Optical scatterometry of subwavelength diffraction gratings: neural-network approach
Applied Optics
|February 21, 2008
Summary
Optical scatterometry uses diffraction patterns to measure diffraction grating geometry. A neural network accurately characterizes subwavelength grating dimensions, offering a promising quantitative analysis method.
Area of Science:
- Metrology
- Optics
- Materials Science
Background:
- Optical scatterometry is an established technique for analyzing diffraction gratings.
- Accurate characterization of grating geometry is crucial for device performance.
Purpose of the Study:
- To demonstrate the efficacy of neural networks for quantitative analysis in optical scatterometry.
- To accurately determine the geometry of subwavelength diffraction gratings.
Main Methods:
- Utilizing optical scatterometry to collect diffraction-pattern data from gratings.
- Employing a neural network model for data analysis and geometric deduction.
Main Results:
- The neural network achieved high accuracy in characterizing grating geometry.
- Specific results include RMS accuracy of 1.9 degrees for groove wall slope, 0.7 nm for linewidth, and 1.0 nm for groove depth.
Conclusions:
- Neural networks provide a powerful and accurate method for quantitative analysis in optical scatterometry.
- This approach enables precise characterization of subwavelength grating dimensions.
