Wafer Parametric Pattern Decomposition for Yield Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Correlating scribe-line test structure measurements to die-level or wafer-level yield is challenging due to the dominance of physical characteristics, making it difficult to isolate and measure specific factors affecting yield, especially when the number of test structures is limited and spatial variations are influenced by multiple factors.
Innovation Solution
The method involves decomposing wafer parametric patterns to remove dominant factors, thereby enhancing less dominant factors, which are then correlated with die-level yield patterns, using techniques like pattern enhancement and machine learning for fault detection and classification, facilitating root cause analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple test structures are designed to isolate specific physical measurements, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the parametric pattern into multiple spatial frequency components through decomposition. By breaking down the complex pattern into distinct spatial frequency bands, the method isolates specific physical factors without requiring multiple separate test structures, thus maintaining measurement precision while reducing device complexity
Solution Approach 2:
The patent applies partial action by selectively removing only the dominant spatial frequency components from the parametric pattern. This partial removal allows the less dominant factors to be revealed without needing to design complete separate test structures for every possible factor, balancing measurement precision with acceptable device complexity
2Device complexity
If the number of test structures is limited, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent transitions from the spatial domain to the spatial frequency domain through decomposition. This dimensional transformation allows a single limited test structure to provide information about multiple physical factors by analyzing different spatial frequency components, thereby maintaining measurement precision with reduced device complexity
Solution Approach 2:
The patent changes the parameter representation from direct spatial measurements to spatial frequency spectrum components. By transforming the measurement parameters through decomposition, the method extracts multiple physical factor information from limited test structure data, preserving measurement precision while minimizing the number of required test structures
3Measurement precision
If dominant factors are removed from parametric pattern, then measurement precision for less dominant factors is improved, but loss of information occurs
Solution Approach 1:
The patent segments the parametric pattern into distinct spatial frequency components and selectively processes them. By separating dominant and less dominant factors into different frequency bands, the method can enhance less dominant factors while preserving the complete information set through the decomposition framework, preventing information loss
Solution Approach 2:
The patent introduces spatial frequency decomposition as an intermediary transformation between the original parametric measurements and the final analysis. This intermediary step allows dominant factors to be removed or reduced for enhanced visibility of less dominant factors, while the complete information can be reconstructed from the decomposed components, avoiding permanent information loss
Data Source
AI summary
Enhancement of less dominant patterns for parametric wafer measurements. Dominant patterns are removed from the parametric pattern thereby revealing a less dominant pattern. The less dominant patterns can be used to identify root causes for yield loss that are not visible in the original parametric measurements.