High performance source optimization using a gradient based method in optical lithography

Yao Peng,  Jinyu Zhang,  Yan Wang,  Zhiping Yu
Tsinghua National Laboratory for Information Science and Technology, Institute of Microelectronics, Tsinghua University


In this paper, we have proposed a highly effective and efficient method to perform source optimization in lithography. Two typical mask patterns, common in the logic and SRAM layout, are adopted in the numerical verifications. Results show that our method can achieve greater image fidelity and over 10X running speed enhancement comparing to Granik’s and Ma&Arce’s methods. Our method is quite robust verified by four different initial source patterns.