Zone-Specific Spectra Libraries for CMP Endpoint Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemical mechanical polishing (CMP) processes face challenges in determining the endpoint of substrate planarization, as variations in substrate zones due to underlying layer thickness and other factors lead to inconsistent material removal rates, making it difficult to accurately determine when a desired flatness or thickness is achieved.
Innovation Solution
The method involves receiving sequences of current spectra of reflected light from different zones of a substrate and comparing them to multiple reference spectra libraries to generate best-match spectra, allowing for zone-specific endpoint determination, thereby compensating for variations in underlying layers and improving endpoint detection reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single reference spectra library is used for all zones, then the system complexity is low, but the endpoint detection reliability deteriorates due to zone variations
Solution Approach 1:
The substrate surface is divided into multiple zones (e.g., center, middle, edge zones) with different radial distances from the center. Each zone has its own reference spectra library tailored to its specific characteristics, allowing endpoint detection to account for zone-specific variations in material removal rates and underlying layer thicknesses.
Solution Approach 2:
Different reference spectra libraries are created for different zones of the substrate, with each library containing reference spectra that are specific to the local characteristics of that zone. This allows the endpoint detection system to use zone-appropriate reference spectra, improving detection reliability by accounting for local variations in polishing behavior.
2Measurement precision
If multiple reference spectra libraries are used for different zones, then the endpoint detection reliability improves, but the device complexity increases
Solution Approach 1:
Reference spectra libraries for all zones are pre-generated and stored in memory before the polishing process begins. Each library is created based on simulated or measured data representing the expected spectral evolution in that specific zone. During polishing, the system simply retrieves and compares against the appropriate pre-prepared library, avoiding the need for real-time complex calculations.
Solution Approach 2:
The system uses reference spectra that are copied from simulated or measured data to create libraries for different zones. These reference spectra represent ideal or target spectral evolution patterns that are stored and used for comparison during actual polishing, allowing the system to detect deviations from expected behavior without requiring complex real-time modeling.
3Manufacturing precision
If zone-specific reference spectra libraries are used, then the thickness uniformity improves, but the processing time increases due to multiple comparisons
Solution Approach 1:
The substrate is divided into discrete zones, each monitored independently with its own reference spectra library. This segmentation allows the system to focus computational resources on comparing spectra within each zone against its specific library, rather than attempting to analyze the entire substrate as a single unit, which would be even more computationally intensive.
Solution Approach 2:
The system monitors spectral parameters that change during polishing and compares them against corresponding parameters in the zone-specific reference libraries. By focusing on key spectral features and parameters rather than analyzing entire spectra in detail, the system can achieve accurate endpoint detection while reducing processing time.
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 reliability of endpoint detection and improves thickness uniformity by using multiple reference spectra libraries tailored to specific zones, reducing the impact of variations in underlying layers and other factors.
Implementation Method 1
receiving a first sequence of current spectra of reflected light from a first zone of a substrate
Data Source
AI summary
A computer-implemented method includes receiving a first sequence of current spectra of reflected light from a first zone of a substrate. A second sequence of current spectra of reflected light from a second zone of the substrate is received. Each current spectrum from the first sequence of current spectra is compared to a plurality of reference spectra from a first reference spectra library to generate a first sequence of best-match reference spectra. Each current spectrum from the second sequence of current spectra is compared to a plurality of reference spectra from a second reference spectra library to generate a second sequence of best-match reference spectra. The second reference spectra library is distinct from the first reference spectra library.