Defect Categorization by Structural Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting and classifying defects in semiconductor wafers are inefficient due to the high volume of defects, making it challenging to identify systematic defects and design-process interactions, which are time and cost prohibitive, and often result in incomplete defect prediction and analysis.
Innovation Solution
A computer-implemented method and system for categorizing defects by determining common structural elements surrounding each defect, using defect clips to match and align structural elements across the wafer, and providing an indication of defects belonging to different structural elements, thereby automating the defect categorization process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection and parsing of individual defects is performed to identify design-process interactions, then defect analysis accuracy is improved, but time consumption and cost increase substantially
Solution Approach 1:
The patent segments the defect analysis process into automated parsing of individual defects, grouping by location patterns, and identification of design-process interactions. This automation resolves the contradiction by performing detailed analysis (improving accuracy) through computational means rather than manual inspection (reducing time consumption).
Solution Approach 2:
The patent replaces the manual mechanical process of inspecting and parsing defects with an automated computer-implemented system. This substitution maintains high defect analysis accuracy while dramatically reducing the time and labor required, resolving the technical contradiction between precision and time loss.
2Reliability
If comprehensive defect inspection is performed across the entire wafer, then defect detection completeness is improved, but data processing complexity increases
Solution Approach 1:
The patent extracts meaningful information from comprehensive defect data by identifying and grouping defects with similar location patterns and characteristics. This extraction approach maintains complete defect detection while simplifying the data processing burden by focusing analysis on grouped patterns rather than individual defects.
Solution Approach 2:
The patent merges individual defect data into grouped defect patterns based on location and characteristics. This combining approach maintains the completeness of comprehensive inspection while reducing processing complexity by analyzing groups rather than individual defects, resolving the contradiction between detection completeness and processing complexity.
3Measurement precision
If repetitive defects are manually correlated into unique defect mechanisms, then defect mechanism identification accuracy is improved, but cost and time requirements become prohibitive
Solution Approach 1:
The patent replaces manual correlation of repetitive defects with an automated computer-implemented system that groups defects by location patterns and identifies underlying mechanisms. This substitution maintains high identification accuracy while dramatically improving productivity by eliminating the prohibitive manual labor requirements.
Solution Approach 2:
The patent segments the complex task of correlating repetitive defects into automated steps: defect detection, location-based grouping, pattern recognition, and mechanism identification. This segmentation enables accurate defect mechanism identification while maintaining high productivity through automation rather than manual processes.
Data Source
AI summary
Methods and apparatus for categorizing defects on workpieces, such as semiconductor wafers and masks used in lithographically writing patterns into such wafers are provided. For some embodiments, by analyzing the layout in the neighborhood of the defect, and matching it to similar defected neighborhoods in different locations across the die, defects may be categorized by common structures in which they occur.