Rigorous modeling and repair of EUV multilayer defects

Buried multilayer (ML) defects in EUV masks continue to pose a significant challenge to imaging fidelity and yield in 0.33 NA EUV lithography. In this work, we utilize the rigorous S-Litho model to simulate the aerial image impact of ML defects across a range of defect depths, lateral offsets, and feature types. The model captures full 3D electromagnetic interactions, enabling accurate prediction of defect printability and through-focus behavior. Building on this characterization capability, we developed a simulation-guided defect repair flow that generates new mask solutions to compensate for reflectivity loss caused by ML defects. The repair strategy is explicitly constrained to be compatible with mask‑repair tool capabilities: supporting removal of absorber material only and accommodating practical repair‑resolution limits. Experimental validation on all selected ML defects from a production EUV mask demonstrated a 100% first‑pass repair success rate using this model-based ML defect repair approach. EUV AIMS measurements confirm that the post‑repair aerial‑image CD closely matches the defect‑free reference, with the worst CD deltas less than 6%.