Basic approaches to simulation of resist mask formation in computational lithography

• Published in: Modern Electronic Materials Vol. 6; no. 1; pp. 37 - 45
• Main Authors: Balan, Nikita N., Ivanov, Vladidmir V., Kuzovkov, Alexey V., Sokolova, Evgenia V., Shamin, Evgeniy S.
• Format: Journal Article
• Language: English
• Published: Moscow Pensoft Publishers 30.03.2020
• Subjects: Calibration; Cluster analysis; Clustering; Contours; image analysis; Lithography; materials; Optimization; Photoresists; Physical simulation; prediction; problem solving; Sampling; Simulation; Topology; Water immersion
• ISSN: 2452-2449; 2452-1779; 2452-1779
• DOI: 10.3897/j.moem.6.1.55056

Main currently used resist mask formation models and problems solved have been overviewed. Stages of "full physical simulation" have been briefly analyzed based on physicochemical principles for conventional diazonapthoquinone (DNQ) photoresists and chemically enhanced ones. We have considered the concepts of the main currently used compact models predicting resist mask contours for full-scale product topologies, i.e., VT5 (Variable Threshold 5) and CM1 (Compact Model 1). Computation examples have been provided for full and compact resist mask formation models. Full resist mask formation simulation has allowed us to optimize the lithographic stack for a new process. Optimal thickness ratios have been found for the binary anti-reflecting layers used in water immersion lithography. VT5 compact model calibration has allowed us to solve the problem of optimal calibration structure sampling for maximal coverage of optical image parameters space while employing the minimal number of structures. This problem has been solved using cluster analysis. Clustering has been implemented using the k -means method. The optimum sampling is 300 to 350 structures, the rms error being 1.4 nm which is slightly greater than the process noise for 100 nm structures. The use of SEM contours for VT5 model calibration allows us to reduce the rms error to 1.18 nm for 40 structures.