Photoresist Parameter Determination via Precomputed Diffraction Signal Library
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for determining photoresist parameters in semiconductor manufacturing are time-consuming and costly due to the need for complex calculations using rigorous coupled wave analysis (RCWA) to generate simulated diffraction signals for quality assurance of periodic gratings.
Innovation Solution
A machine learning system is trained to generate simulated diffraction signals using photoresist parameters, which are then associated with profile parameters to characterize geometric features, allowing for efficient determination of photoresist behavior and structure profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rigorous coupled wave analysis (RCWA) is used to generate simulated diffraction signals, then measurement precision is improved, but loss of time increases and device complexity increases
Solution Approach 1:
The patent pre-calculates and stores a library of diffraction signals corresponding to various photoresist parameters before actual measurement. When a measured diffraction signal is obtained, the system quickly searches and compares it against the pre-computed library to determine photoresist parameters, avoiding time-consuming real-time RCWA calculations while maintaining measurement precision.
Solution Approach 2:
The patent creates a library of copied diffraction signals from RCWA calculations for different photoresist parameter values. Instead of performing new RCWA calculations during measurement, the system copies and compares the measured signal against these pre-generated signal copies, significantly reducing computational time while preserving accuracy.
2Measurement precision
If rigorous coupled wave analysis (RCWA) is used to generate simulated diffraction signals, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The complex RCWA calculations are performed in advance and the results are stored in a library. The actual measurement system only needs to perform simple signal comparison and parameter lookup, dramatically reducing the computational complexity of the measurement device while maintaining high measurement precision through the use of pre-computed reference signals.
Solution Approach 2:
The patent replaces complex real-time RCWA calculations with copying and comparing pre-generated diffraction signal patterns. This substitution reduces the computational complexity from solving Maxwell's equations in real-time to simple pattern matching, making the measurement system more practical while preserving measurement accuracy.
3Productivity
If machine learning is used to generate simulated diffraction signals, then productivity is improved, but manufacturing precision may be compromised
Solution Approach 1:
The system performs preliminary RCWA calculations to generate a comprehensive library of diffraction signals covering the full range of expected photoresist parameters. This pre-computation ensures that the machine learning model is trained on accurate reference data, enabling fast parameter determination while maintaining manufacturing precision through comparison against rigorously calculated reference signals.
Solution Approach 2:
The patent introduces a library of pre-calculated diffraction signals as an intermediary between the complex RCWA method and the fast machine learning approach. The machine learning model learns the relationship between measured signals and photoresist parameters from this intermediary library, combining the accuracy of RCWA with the speed of machine learning while preserving manufacturing precision.
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
This approach reduces computational costs and time by using machine learning to generate and match diffraction signals, enabling faster and more accurate determination of photoresist parameters and structure profiles in semiconductor manufacturing.
Implementation Method 1
The periodic grating is then illuminated with an electromagnetic radiation. The electromagnetic radiation that deflects off of the periodic grating are collected as a diffraction signal.
Data Source
AI summary
To generate a simulated diffraction signal, one or more values of one or more photoresist parameters, which characterize behavior of photoresist when the photoresist undergoes processing steps in a wafer application, are obtained. One or more values of one or more profile parameters are derived using the one or more values of the one or more photoresist parameters. The one or more profile parameters characterize one or more geometric features of the structure. A simulated diffraction signal is generated using the one or more values of the one or more profile parameters. The simulated diffraction signal characterizes behavior of light diffracted from the structure. The generated simulated diffraction signal is associated with the one or more values of the one or more photoresist parameters. The generated simulated diffraction signal, the one or more values of the one or more photoresist parameters, and the association between the generated simulated diffraction signal and the one or more values of the one or more photoresist parameters are stored.