Optimizing extreme ultraviolet lithography imaging metrics as a function of absorber thickness and illumination source: a simulation case study of Ta-Co alloy

Abstract

Background: The optical properties of the absorber material and its thickness, along with the choice of illumination, govern the imaging performance of an EUV photomask. Consequently, the imaging metrics, viz., normalized image log slope (NILS), telecentricity error (TCE), and best focus variation (BFV) through pitch, exhibit a trade-off. Previous studies have focused on either reflectivity or phase shift induced by the absorber to determine the optimum absorber thickness, which ignores the structure of the patterns on the mask.Aim: This simulation study intends to facilitate the optimum absorber thickness selection with an emphasis on diffraction order analysis of the line and space (LnS) pattern and investigates the impact of illumination source shapes for a Ta-Co alloy absorber mask.Approach: We investigate the behavior of imaging metrics as a function of the absorber thickness using literature-recommended illumination source shapes. NILS, TCE, throughput criterion, diffraction order's amplitude ratio, and phase difference are the critical parameters when selecting the optimum absorber thickness. A through-pitch imaging performance of the Ta-Co alloy absorber with recommended thicknesses is compared with the reference 60 nm Ta-based absorber using leaf dipole (LDP), inner half leaf dipole (IHLDP), and outer half leaf dipole (OHLDP) for LnS patterns with a target trench of 10 nm and pitch ranging from 20 to 40 nm.Results: The proposed absorber thickness optimization methodology validates the reference Ta-based absorber thickness and recommends 52 nm of Ta-Co alloy absorber thickness. IHLDP and OHLDP demonstrate enhanced NILS for the smallest pitch. However, this behavior is inconsistent throughout the pitch. IHLDP exhibits low TCE. However, because of higher BFV, the advantages are not exceptionally beneficial.Conclusions: A holistic approach makes the proposed absorber thickness optimization methodology reliable and generally applicable. The through-pitch simulation outcomes at selected absorber thickness favor a conventional full-leaf dipole for the studied imaging metrics.(c) 2023 Society of Photo-Optical Instrumentation Engineers (SPIE)