Semiconductor Defect Offset Correction for Tool Coordinate Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The misalignment between different examination tools' coordinate systems in semiconductor manufacturing leads to inefficiencies in defect detection, requiring manual intervention and reducing throughput, especially as semiconductor devices become more complex with shrinking feature dimensions.
Innovation Solution
An automated system using a trained machine learning model to rank defect candidates by probability and calculate an offset between inspection and review coordinate systems, enabling efficient alignment without manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual intervention is used to align coordinate systems between examination tools, then alignment accuracy can be achieved, but productivity decreases and time is lost
Solution Approach 1:
The system performs automatic offset calculation using defect candidate locations as reference points. The examination tool itself generates the alignment data by comparing defect locations between coordinate systems, eliminating the need for external manual intervention while maintaining alignment accuracy
Solution Approach 2:
The system pre-calculates the offset between coordinate systems using defect candidates identified during the examination process. This preliminary alignment calculation enables subsequent reviews to proceed without repeated manual alignment steps, improving throughput while maintaining accuracy
2Adaptability or versatility
If FOV is enlarged to accommodate coordinate misalignment, then all defect candidates can be reviewed, but review time increases and productivity decreases
Solution Approach 1:
Instead of uniformly enlarging the FOV across the entire examination area, the system applies local coordinate offset correction specifically to defect candidate locations. This targeted approach maintains review coverage for all defects while avoiding the time penalty of examining unnecessary areas with enlarged FOV
Solution Approach 2:
The offset calculation is performed in advance using defect candidate locations, enabling the review tool to directly navigate to corrected defect positions without requiring FOV enlargement. This preliminary correction maintains both review coverage and efficiency
3Measurement precision
If manual defect selection is performed to establish coordinate offset, then alignment can be achieved, but the process becomes complex and time-consuming
Solution Approach 1:
The system automatically identifies defect candidates and uses their locations to calculate coordinate offsets. The examination tool self-performs the alignment function that previously required manual operator intervention, simplifying the process while maintaining precision
Solution Approach 2:
Defect candidates are automatically identified and their locations are pre-processed to calculate the offset between coordinate systems. This automated preliminary action eliminates the complex manual steps of defect selection and alignment establishment
Data Source
AI summary
There is provided a system and method of examination of a semiconductor specimen. The method includes obtaining a group of defect candidates associated with respective inspection locations represented in an inspection coordinate system; using a trained machine learning (ML) model to provide, for each defect candidate, a probability of the defect candidate being a defect of interest (DOI), and ranking the group of defect candidates to an ordered list according to respective probabilities thereof, in response to a part of the ordered list of defect candidates being reviewed by a review tool in accordance with an order thereof, receiving a predefined number of DOIs associated with respective review locations represented in a review coordinate system; and calculating an offset between the review coordinate system and the inspection coordinate system based on respective inspection and review locations associated with the predefined number of DOIs.