Optical Measuring Apparatus for Thin Layer Thickness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical measuring methods face challenges in precisely measuring the thickness of thin layers on patterns due to correlations with underlying structures, particularly in semiconductor manufacturing where precise thickness measurement of nano-scale features is crucial.
Innovation Solution
An optical measuring method utilizing spectroscopic ellipsometry that involves obtaining first and second spectra from a pattern and a thin layer, calculating a skew spectrum, and performing a Fourier transform operation to determine the thickness of the thin layer, with the option to use reference data for layers of different thicknesses and applying modeling techniques to calculate pattern profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional spectroscopic ellipsometry is used to measure thin layer thickness, then the measurement process is simple, but the measurement precision deteriorates due to correlation with underlying structures
Solution Approach 1:
The patent segments the measurement process into distinct phases: obtaining a first spectrum from the underlying pattern, obtaining a second spectrum from the thin layer on the pattern, calculating a skew spectrum between them, and performing Fourier transform. This segmentation separates the thin layer measurement from the underlying structure correlation, enabling precise thickness measurement independent of pattern complexity.
Solution Approach 2:
The skew spectrum acts as an intermediary that isolates the thin layer information from the underlying pattern. By calculating the difference or ratio between the first spectrum (pattern only) and second spectrum (pattern + thin layer), the method creates an intermediate representation that contains only the thin layer's optical properties, eliminating correlation with the underlying structure.
2Measurement precision
If conventional ellipsometry measures thin layers on patterns, then the process is straightforward, but the measurement precision deteriorates due to noise from underlying structures
Solution Approach 1:
The method extracts the thin layer signal from the total measurement by obtaining spectra at two different states (with and without thin layer) and calculating a skew spectrum. This extraction process removes the underlying pattern's contribution and isolates only the thin layer's optical signal, eliminating noise from the underlying structures.
Solution Approach 2:
The patent converts the harmful correlation with underlying structures into a beneficial measurement approach. By intentionally measuring the pattern both with and without the thin layer, and then calculating the skew spectrum, the method transforms the interfering pattern signal into a reference that enables precise thin layer measurement through subtraction or ratio operations.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables precise measurement of thin layer thicknesses down to several angstroms, effectively addressing the complexity of underlying structures like grating patterns, and providing accurate thickness information while removing noise components.
Implementation Method 1
obtaining an amplitude ratio (tan(Ψ) or phase difference (Δ) spectrum of a reflected light from the pattern and the thin layer on the pattern
Implementation Method 2
performing a fourier transform operation on the skew spectrum to obtain a fourier transform spectrum, and determining the thickness of the thin layer based on a peak position of the fourier transform spectrum
Data Source
AI summary
In an optical measuring method, a first spectrum and a second spectrum are obtained from a pattern and a thin layer formed on the pattern by a deposition process using an ellipsometer respectively. A skew spectrum is obtained between the first spectrum and the second spectrum. A fourier transform operation is performed on the skew spectrum to calculate a thickness of the thin layer on the pattern.