Dynamic Metrology Sampling Plan for Semiconductor Wafer Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current metrology sampling plans are not sensitive enough to subtle variations in semiconductor wafer processes, often missing critical excursions, especially at technology nodes beyond 65 nm, due to their fixed nature and limited coverage.
Innovation Solution
A computer-implemented method generates a dynamic metrology sampling plan by identifying individual defects with abnormal attributes within a predetermined pattern on the wafer, focusing sampling on areas where these defects are located, using statistical analysis and design-aware inspection data to create a plan that adapts to process variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If fixed metrology sampling locations are used, then measurement speed is improved, but measurement precision deteriorates due to inability to detect subtle process variations
Solution Approach 1:
The patent transforms the static, fixed metrology sampling locations into dynamic locations that adapt based on inspection results. The system identifies abnormal defects during inspection and automatically adjusts metrology sampling locations to target these abnormal areas, enabling the sampling plan to respond to actual process variations rather than following a predetermined fixed pattern.
Solution Approach 2:
The system establishes a feedback loop where inspection results (including defect detection and abnormal attribute identification) inform and adjust subsequent metrology sampling locations. This closed-loop approach allows the system to use information from inspection to optimize where metrology measurements are performed, improving detection capability for subtle variations while maintaining efficient sampling rates.
2Device complexity
If fixed metrology sampling locations are used, then device complexity is reduced, but reliability deteriorates due to missing critical excursions
Solution Approach 1:
The sampling plan transitions from a static, pre-defined location set to a dynamic system that automatically adjusts sampling locations based on real-time inspection data. This dynamic adaptation enables the system to reliably detect critical excursions and abnormal process variations without requiring complex manual configuration or intervention.
Solution Approach 2:
The system performs self-adjustment by automatically identifying abnormal defects during inspection and autonomously determining optimized metrology sampling locations based on these findings. This self-service capability eliminates the need for external manual planning while improving reliability through adaptive targeting of critical areas.
3Productivity
If metrology is performed at limited locations, then productivity is improved, but measurement precision deteriorates due to insufficient wafer coverage
Solution Approach 1:
Instead of uniformly sampling across the entire wafer or using fixed locations, the system applies local quality by concentrating metrology measurements at specific locations where abnormal defects are detected during inspection. This targeted approach ensures that limited metrology resources are allocated to the most critical areas of the wafer, maximizing the informational value of each measurement while maintaining productivity.
Solution Approach 2:
The system uses feedback from inspection results to dynamically determine which locations require metrology measurement. By identifying abnormal defects and their locations during inspection, the system directs metrology resources to these specific areas, ensuring that limited measurements provide maximum insight into wafer characteristics and process variations.
Data Source
AI summary
Various computer-implemented methods, carrier media, and systems for generating a metrology sampling plan are provided. One computer-implemented method for generating a metrology sampling plan includes identifying one or more individual defects that have one or more attributes that are abnormal from one or more attributes of a population of defects in which the individual defects are included. The population of defects is located in a predetermined pattern on a wafer. The method also includes generating the metrology sampling plan based on results of the identifying step such that one or more areas on the wafer in which the one or more identified individual defects are located are sampled during metrology.