Pattern Inspection via Curve Fitting for Fine EUV Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pattern inspection methods, particularly for fine patterns formed using extreme ultra-violet (EUV) patterning techniques, suffer from low accuracy in measuring critical dimensions and optical proximity correction (OPC) patterns in semiconductor manufacturing.
Innovation Solution
A pattern inspection method involving image processing techniques such as contour extraction, curve-fitting using functions like Sigmoid, hyperbolic tangent, and Fermi-Dirac, and data analysis to generate pattern inspection data including width, height, and slope, which improves the accuracy of pattern measurement and consistency analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a scanning electron microscope (SEM) is used to measure critical dimensions of fine patterns, then the measurement capability is provided, but the measurement precision is low for highly fine patterns formed by EUV patterning technique
Solution Approach 1:
The patent transforms the raw contour data into enhanced measurement parameters through curve fitting using Sigmoid, hyperbolic tangent, or Fermi-Dirac functions. This mathematical transformation extracts precise dimensional information (width, height, slope) from the contour data, significantly improving measurement precision for fine EUV patterns beyond what traditional SEM measurement can achieve
Solution Approach 2:
The patent introduces contour extraction as an intermediary step between image acquisition and measurement. By first extracting the contour of the pattern and then performing curve fitting on the contour data, the system creates an intermediate representation that enhances the measurability of fine patterns, effectively bridging the gap between SEM imaging capabilities and precise measurement requirements
2Measurement precision
If traditional image processing is used for pattern inspection, then the inspection process is simple, but the inspection accuracy is insufficient for fine patterns
Solution Approach 1:
The patent segments the pattern inspection process into distinct stages: contour extraction, curve fitting, and parameter calculation. This segmentation allows each stage to be optimized independently, with contour extraction handling edge detection and curve fitting handling precise dimensional extraction, thereby improving overall inspection accuracy while maintaining manageable complexity through modular processing
Solution Approach 2:
The patent performs preliminary contour extraction and curve fitting operations before final measurement calculation. By pre-processing the image data to extract contours and fit curves in advance, the system prepares enhanced measurement data that can be directly used for accurate pattern inspection, improving accuracy while organizing complexity into preparatory and final analysis stages
Data Source
AI summary
Provided is a pattern inspection method including obtaining an image of a substrate on which a pattern is formed, extracting a contour based on the image, detecting positions of a target pattern based on the contour, generating pattern inspection data by performing a curve-fitting on the detected positions of the target pattern, and analyzing the pattern based on the pattern inspection data, wherein the curve-fitting is performed by using at least one of a Sigmoid function, a hyperbolic tangent function, and a Fermi-Dirac function, and wherein the pattern inspection data includes a width in a first direction, a height in a second direction, and a pattern slope of the target pattern.