Plasma Etching Endpoint Detection Using PCA and OES Wavelength Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor device manufacturing, accurately stopping the plasma etching process without etching the lower layer is challenging due to the difficulty in distinguishing the chemical composition of the upper and lower layers using existing optical emission spectrometry methods.
Innovation Solution
A substrate processing method that selects specific wavelengths with high correlation to the etching endpoint, applies principal component analysis and Gaussian mixture models to preprocess and cluster optical emission spectrometry data, and uses these models to determine the endpoint of the etching process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical emission spectrometry is used to monitor chemical properties of gas in plasma processing chamber, then endpoint detection capability is improved, but measurement precision deteriorates due to difficulty in distinguishing chemical composition of upper and lower layers
Solution Approach 1:
The patent segments the optical emission spectrometry data by selecting specific wavelength bands that correspond to different chemical species. Instead of analyzing the entire spectrum, the method divides the wavelength range into multiple bands and selects those with high correlation to the endpoint, thereby improving the ability to distinguish between upper and lower layer chemical compositions.
Solution Approach 2:
The patent transforms the spectral data from wavelength-intensity space to a reduced-dimensional space using principal component analysis. This dimensionality reduction projects the high-dimensional spectral data onto a lower-dimensional manifold while preserving the variance information, enabling better separation of overlapping spectral signatures from different layers.
2Reliability
If multiple wavelength data are collected for endpoint detection, then detection reliability is improved, but device complexity increases due to data processing requirements
Solution Approach 1:
The patent extracts only the most relevant information from the multiple wavelength data by selecting specific wavelength bands with high correlation to the endpoint. This extraction process removes redundant and irrelevant wavelength information, reducing the data processing burden while maintaining detection reliability.
Solution Approach 2:
The patent changes the parameter representation of the spectral data by applying principal component analysis to transform wavelength-intensity data into principal component scores. This parameter transformation simplifies the data structure and reduces complexity while preserving the essential variations related to endpoint detection.
3Productivity
If dimensional reduction is applied to wavelength data, then processing efficiency is improved, but information loss may occur
Solution Approach 1:
The patent performs preliminary wavelength selection before dimensional reduction by identifying and selecting only those wavelength bands with high correlation to the endpoint. This preliminary action ensures that the subsequent dimensional reduction operates on already-filtered, relevant data, minimizing information loss while maintaining processing efficiency.
Solution Approach 2:
The patent uses feedback from the correlation analysis to guide the wavelength selection process. By calculating correlation between wavelength data and endpoint timing, the system provides feedback to identify which wavelengths contain the most useful information, ensuring that dimensional reduction retains critical spectral features.
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 method improves the reliability of endpoint detection in the etching process by reducing noise and dimensionality, allowing for precise termination of the etching process based on deviations in labeled data over time, thus preventing over-etching of the lower layer.
Implementation Method 1
optical emission spectrometry (OES) may be used to monitor the chemical properties of the gas in the plasma processing chamber
Implementation Method 2
a plasma source configured to generate plasma for processing a process substrate within the chamber
Data Source
AI summary
A substrate processing method includes collecting a plurality of pieces of optical emission spectrometry data including a wavelength, intensity of the wavelength, and time using optical emission spectrometry on a plurality of substrates, selecting a selected wavelength band having a high correlation with an endpoint of an etching process from the plurality of pieces of optical emission spectrometry data, preprocessing the plurality of pieces of optical emission spectrometry data to generate a selected dataset, generating a principal component analysis model using the selected dataset, generating a probability distribution model capable of clustering data of the principal component analysis model, and performing the etching process on a process substrate using the principal component analysis model and the probability distribution model.