Defect Observation Device Adaptive Cell Die Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor manufacturing, the increased complexity of design patterns and manufacturing processes lead to diversified defects, making it difficult to set optimal observation conditions for defect detection, resulting in reduced throughput and accuracy, especially when transitioning between cell and die comparison methods.
Innovation Solution
A defect detection processing mode is determined using a reference image to accurately detect defects, optimizing conditions for either cell or die comparison methods based on the pattern's periodicity, thereby stabilizing throughput and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cell comparison method is used for defect detection, then throughput is improved, but defect detection accuracy deteriorates when defects cannot be detected by cell comparison
Solution Approach 1:
The system dynamically switches between cell comparison method and die comparison method based on the detection results. When cell comparison fails to detect a defect, the system transitions to die comparison method, making the detection process adaptive rather than static. This resolves the contradiction by allowing the system to prioritize throughput with cell comparison while ensuring accuracy through conditional die comparison.
Solution Approach 2:
The system uses feedback from the cell comparison detection results to determine whether to proceed to die comparison. The feedback mechanism evaluates whether a defect was detected and triggers the appropriate method transition, ensuring that throughput is maximized when possible while maintaining detection accuracy through fallback mechanisms.
2Measurement precision
If die comparison method is used for defect detection, then defect detection accuracy is improved, but throughput deteriorates due to reference image acquisition
Solution Approach 1:
The system applies die comparison method partially, only when cell comparison fails to detect a defect. Instead of applying die comparison to all cases (which would maximize accuracy but minimize throughput), the system uses it selectively as a supplement to cell comparison, achieving the optimal balance between accuracy and throughput.
3Adaptability or versatility
If multiple defect detection methods are prepared for different purposes, then adaptability is improved, but device complexity increases
Solution Approach 1:
The system implements a universal defect detection framework that can perform both cell comparison and die comparison methods through a single integrated system. The control unit manages multiple detection methods within one device, eliminating the need for separate dedicated systems for each method and reducing overall device complexity while maintaining versatility.
Data Source
AI summary
In a scheme for analyzing low magnification defect images and determining whether or not a defect detection method using cell comparison is applicable, if a defect detection method using cell comparison cannot be applied and the proportion transitioning to a defect detection method using die comparison increases, throughput may decrease even more than starting out with defect detection by a defect detection method using die comparison. The purpose of present invention is to carry out high precision defect detection with a stable throughput. In the present invention, the defect detection processing mode applied for detecting defects from the defect image is determined using a reference image, and defects are detected from the defect image by the defect detection processing mode that has been determined.