FDTD-Based Study on Equivalent Medium Approximation Model of Surface Roughness for Thin Films Characterization Using Spectroscopic Ellipsometry

Spectroscopic ellipsometry (SE) is an effective method to measure the optical constants of thin film materials which is very sensitive to the surface topography of thin films. When performing ellipsometric measurements of the optical constants of solid materials with rough surfaces, the equivalent m...

Full description

Bibliographic Details
Main Authors: Wanpei Yu, Changcai Cui, Huihui Li, Subiao Bian, Xi Chen
Format: Article
Language:English
Published: MDPI AG 2022-08-01
Series:Photonics
Subjects:
Online Access:https://www.mdpi.com/2304-6732/9/9/621
Description
Summary:Spectroscopic ellipsometry (SE) is an effective method to measure the optical constants of thin film materials which is very sensitive to the surface topography of thin films. When performing ellipsometric measurements of the optical constants of solid materials with rough surfaces, the equivalent medium approximation (EMA) model is often used to characterize the surface topography. The EMA model is determined by two parameters of equivalent thickness <i>d</i><sub>EMA</sub> and the void volume fraction <i>f</i>. In most applications, the void volume fraction parameter <i>f</i> is always set to an empirical 50% without any instructions, and then the thickness parameter <i>d</i><sub>EMA</sub> is determined by fitting. In order to improve the accuracy of the fitting results, it is necessary to validate the construction law of the EMA model in the ellipsometry analysis considering characteristic parameters of the actual surface topography. In this paper, the influence of the surface topographical parameters on EMA model is analyzed. The method of FDTD (finite difference time domain) is employed to simulate the SiO<sub>2</sub> films with different topographical parameters and EMA model are carried out on these samples. The analysis results show that the EMA model constructed with <i>d</i><sub>EMA</sub> = <i>σ</i> + 0.80 <i>h</i> (<i>σ</i>: the root mean square height, <i>h</i>: the average height) can better fit the SE parameters. The proposed method can facilitate a better understanding and utilization of the EMA model in SE application.
ISSN:2304-6732