Extreme ultraviolet (EUV) multilayer film high-precision representation method based on dual-target genetic algorithm

Abstract

The present invention discloses an extreme ultraviolet multilayer film high-precision representation method based on a dual-target genetic algorithm. The method is characterized by applying a real number encoded non-dominated sorting genetic algorithm (NSGA-II) to the joint fitting of the experiment results of a grazing incidence X-ray reflectance spectrum and an EUV reflectance spectrum of an EUV multilayer film, taking the grazing incidence X-ray reflectance spectrum and the EUV reflectance spectrum of the EUV multilayer film as the optimization targets of the dual-target genetic algorithm, evolving to obtain a non-dominated solution set approaching a Pareto front, utilizing a Levnberg-Marquart algorithm to further optimize the superior individual having the smaller fitting residual in two fitting targets in the non-dominated solution set, and obtaining an optimal structural parameter of the smallest fitting residual of the two fitting targets. The method of the present invention solves a multi-solution problem based on the single target fitting (the grazing incidence X-ray reflectance spectrum or the EUV reflectance spectrum) solution, avoids the problem that when the two fitting targets are added simply and fitted jointly to solve, the targets influence mutually, even one target is influenced severely, so that the representation precision of a multilayer film microstructure is not high.

Description

technical field [0001] The invention specifically relates to a high-precision characterization method for the EUV multilayer film microstructure required by the multilayer film mirror used in extreme ultraviolet (EUV) lithography technology. Background technique [0002] EUV lithography technology is considered to be the most promising next-generation lithography technology to meet the needs of the semiconductor industry. But in the EUV band, almost all materials are opaque, and the refractive index is very close to 1, so the EUV optical system cannot use traditional refractive optical components, but must use reflective optical systems. Therefore, the multilayer film that achieves high reflectivity of EUV light has become the core optical component of the EUV optical system. At the same time, the EUV multilayer film has also become the hot spot and core of technological research and development in the field of EUV optics, and has attracted widespread attention from domestic...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems existing in the high-precision characterization of the microstructure of the existing EUV multilayer film, the present invention provides a high-precision characterization method for EUV multilayer films based on a dual-objective genetic algorithm. The GIXR and EUV reflectance spectra of periodic EUV multilayer films, through the fitting solution and evolution of the double-objective genetic algorithm, obtain the microstructural parameters of EUV multilayer films with high precision, and solve the problem of GIXR or EUV reflectance spectroscopy) has the problem of low characterization accuracy

Images