Azimuth-guided image fusion method for sub-λ/18 defect inspection using a laser scattering darkfield imaging system

Area of Science:

• Semiconductor manufacturing
• Optical metrology
• Nanotechnology

Background:

• Continuous scaling of advanced process nodes presents manufacturing challenges.
• High-sensitivity and high-throughput defect inspection methods are in demand.
• Visible spectrum darkfield imaging lacks nanoscale sensitivity due to noise and surface roughness.

Purpose of the Study:

• To develop a rigorous physical model for defect imaging.
• To propose an azimuth-guided image fusion strategy for enhanced defect detection.
• To overcome the sensitivity limitations of darkfield imaging in the visible spectrum.

Main Methods:

• Constructed a rigorous physical model based on vectorial electromagnetic theory.
• Developed an azimuth-guided image fusion strategy exploiting defect scattering dependence on illumination azimuth.
• Configured an optimal laser scattering darkfield imaging system.

Main Results:

• Systematically elucidated mechanisms affecting defect detectability.
• Achieved effective identification of sub-wavelength (sub-λ/18) defects.
• Demonstrated enhanced detection capability for randomly oriented subwavelength defects.

Conclusions:

• The azimuth-guided fusion method overcomes sensitivity limitations in visible spectrum darkfield imaging.
• This work provides an efficient solution for nanoscale defect inspection at advanced semiconductor nodes.
• The developed physical model aids in understanding and improving defect detection mechanisms.