Robust hybrid source and mask optimization to lithography source blur and flare.

Area of Science:

• Semiconductor manufacturing
• Photolithography
• Optical engineering

Background:

• Source and Mask Optimization (SMO) is crucial for advanced lithography nodes (45 nm and beyond).
• Scanner errors, specifically source blur and flare, significantly degrade the performance of traditional SMO-optimized sources and masks.
• Existing SMO methods lack effective compensation for these non-ideal scanner factors.

Purpose of the Study:

• To develop a robust hybrid SMO (HSMO) method that compensates for scanner errors.
• To integrate sensitivities to source blur and flare into the cost function for improved robustness.
• To enhance pattern fidelity and process window in advanced lithography.

Main Methods:

• Developed a hybrid Source and Mask Optimization (HSMO) approach.
• Incorporated sensitivities of the aerial image to source blur and flare into the cost function.
• Performed simulations comparing HSMO with traditional SMO.

Main Results:

• The proposed source blur-flare-aware HSMO method demonstrates superior performance compared to traditional SMO.
• HSMO effectively mitigates the negative impacts of scanner errors on lithography outcomes.
• Significant improvements in pattern fidelity and process window were observed with HSMO.

Conclusions:

• The developed HSMO method offers a robust solution for advanced lithography challenges posed by scanner imperfections.
• Integrating error sensitivities into the optimization process is key to achieving high-fidelity patterns and wider process windows.
• This approach is vital for reliable semiconductor manufacturing at critical nodes.