Fast optimization of defect compensation and optical proximity correction for extreme ultraviolet lithography mask

Abstract

Mask multilayer defect and optical proximity effect bring harmful impact on the imaging quality of extreme ultraviolet lithography. The undesirable impact could be compensated for by modification of the mask patterns. Based on machine learning and quasi-rigorous electromagnetic method, two fast simulation methods are utilized to accelerate the optimization of the defect compensation patterns. Defect compensation considering simultaneous optical proximity correction is also studied for further improvement of pattern fidelity. Combined with a well-performed evolutionary searching algorithm, the two fast simulation methods are adopted in the defect-compensation optimization for different mask patterns. Mask modification optimization considering both defect compensation and optical proximity correction is also demonstrated by a simulation study. Simulation results demonstrated the acceleration due to the adoption of the two fast simulation methods. Simulation results also validated that both the defect impact and the optical proximity effect can be compensated for under the same mask modification frame. The mask modification strategy combining the two fast simulation methods and the evolutionary algorithm provides a potential and flexible compensation approach for EUVL masks that influenced by multilayer defect and proximity effect.