Optical Spectra Endpoint Detection in CMP
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemical mechanical polishing (CMP) processes face challenges in determining the endpoint of substrate planarization, leading to variations in material removal rates due to initial thickness, slurry composition, polishing pad conditions, and other factors, resulting in potential overpolishing or underpolishing issues that affect circuit performance.
Innovation Solution
The implementation of an optical detector system that obtains spectra from the substrate during polishing and compares them to a library using techniques like least sum of squares matching and robust line fitting to determine the polishing endpoint, allowing for accurate identification of the target thickness and preventing outlier spectra from skewing results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If polishing time is used to determine endpoint, then process simplicity is maintained, but manufacturing precision deteriorates due to variations in material removal rate
Solution Approach 1:
The patent replaces mechanical/time-based endpoint determination with optical spectroscopy. Instead of relying on polishing time and mechanical measurements, the system uses reflectance spectra analysis to detect the polishing endpoint, achieving both precision and automation without complex mechanical intervention.
Solution Approach 2:
The patent introduces spectra as an intermediary parameter between the polishing process and endpoint determination. By monitoring changes in reflectance spectra during polishing, the system indirectly detects the endpoint condition, allowing for precise control without direct mechanical measurement of the polished surface.
2Manufacturing precision
If spectra analysis is implemented for endpoint determination, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The polishing system is equipped with a multi-functional optical detection device that can perform multiple operations: acquiring reflectance spectra, processing the spectral data, comparing against reference spectra, and determining the endpoint. This universal device consolidates what could be multiple separate systems into one integrated unit, reducing overall complexity.
Solution Approach 2:
The system performs self-diagnosis and self-determination of the polishing endpoint through automated spectra analysis. The device automatically compares real-time spectra with reference spectra and determines when polishing should stop, eliminating the need for manual intervention or additional complex control systems.
3Productivity
If automated spectra-based endpoint detection is used, then productivity is improved through precise control, but loss of information increases due to potential outliers in spectral data
Solution Approach 1:
The system performs preliminary actions by acquiring multiple spectra during the polishing process and identifying outliers before making the final endpoint determination. By pre-processing the spectral data to remove outliers and validate measurements, the system ensures accurate endpoint detection without sacrificing productivity.
Solution Approach 2:
The system implements feedback mechanisms where spectra are continuously monitored and compared against reference data. The endpoint determination is based on feedback from the spectral analysis, allowing the system to adjust and confirm the endpoint decision, ensuring that no critical information is lost in the automated process.
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 enables precise determination of the polishing endpoint, reducing variations in material removal and improving the accuracy of planarization, thereby preventing electrical shorts and increased circuit resistance.
Implementation Method 1
an optical detector is used to obtain spectra from a substrate being polished
Data Source
AI summary
Methods of determining a polishing endpoint are described using spectra obtained during a polishing sequence. In particular, techniques for using only desired spectra, faster searching methods and more robust rate determination methods are described.