Spectral Data Dimensional Reduction for CMP Endpoint Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemical mechanical polishing (CMP) processes face challenges in determining the polishing endpoint due to variations in material removal rates caused by initial substrate thickness, slurry composition, polishing pad conditions, and relative speed, leading to inaccuracies in measuring wafer thickness and uniformity.
Innovation Solution
A method involving spectral data processing using singular value decomposition, CUR matrix approximation, or principal component analysis to reduce noise in reflected spectra, allowing for precise measurement of wafer thickness and improved endpoint detection in CMP processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If optical monitoring is performed in-situ during polishing, then real-time endpoint detection is improved, but measurement precision deteriorates due to noise in spectral data
Solution Approach 1:
The patent extracts and removes noise components from spectral data through signal processing techniques. By separating the useful spectral information from noise contaminants, the system maintains real-time monitoring capability while improving measurement precision of wafer thickness during polishing.
Solution Approach 2:
The system implements feedback by continuously monitoring spectral data during polishing, processing the signals to remove noise, and using the cleaned spectral information to adjust and control the polishing process in real-time, thereby improving both response time and measurement accuracy.
2Adaptability or versatility
If material removal rate variations are accommodated, then adaptability is improved, but manufacturing precision deteriorates due to wafer-to-wafer non-uniformity
Solution Approach 1:
The patent applies local quality by measuring and analyzing spectral data from different locations on the wafer surface. This enables detection of local thickness variations and non-uniformity across the wafer, allowing targeted corrections to achieve improved manufacturing precision while maintaining adaptability to overall process variations.
Solution Approach 2:
The system performs preliminary measurements and establishes baseline spectral characteristics before polishing begins. This preliminary action enables the system to compensate for expected variations in material removal rate and predict endpoint conditions, thereby maintaining precision despite process adaptability requirements.
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 enhances the precision of wafer thickness measurement and reduces wafer-to-wafer non-uniformity, improving the reliability of polishing endpoint detection.
Implementation Method 1
a spectrum of reflected light from the substrate is measured
Data Source
AI summary
A plurality of spectra reflected from one or more substrates at a plurality of different positions on the one or more substrates are represented in the form of a first matrix, and the first matrix is decomposed into products of at least two component matrixes of a first set of component matrixes. The dimensions of each of the at least two component matrixes is reduced to produce a second set of component matrixes containing the at least two matrixes with reduced dimensions.