Calibration of Gaussian Random Field stochastic EUV models

Abstract

Typical photoresists processes include a small set of photo-chemical reactions, with each reaction represented by many statistically identically distributed and independent instances. These instances eventually combine into the resist deprotection function, resulting, by virtue of the Central Limit Theorem, in Gaussian Random Field deprotection models. We discuss and demonstrate the approaches to calibration of such models, based on experimentally measured edges of lithographic features, their LER, LWR and PSD. We also present, discuss and analyze the phenomenon of "spatial ergodicity" and its effect on proper sampling of edge measurements for stochastic model calibration.