Metrology Target Characterization via Geometric Abnormality Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current metrology methods face challenges in accurately identifying and classifying abnormalities in metrology targets on wafers, which are crucial for error quantification and process control in semiconductor manufacturing, as they often rely on complex algorithms and are prone to noise and outlier issues.
Innovation Solution
The development of a system that derives target signals from wafer targets, calculates specific metrics using respective functions, and analyzes these metrics to characterize targets, allowing for the geometric classification of abnormalities and linking them to corresponding error sources, thereby enhancing metrology measurements and process control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current metrology methods are used to measure overlay between produced layers, then measurement can be performed, but accuracy in identifying and classifying metrology target abnormalities is insufficient
Solution Approach 1:
The patent segments the metrology measurement process into distinct functional modules: target signal derivation module, metric calculation module, and target characterization module. Each module performs a specific function (deriving signals from targets, calculating metrics from signals, analyzing metrics to characterize targets), which improves measurement precision while managing system complexity through modular design
Solution Approach 2:
The patent introduces metrics as intermediary variables between raw target signals and final target characterization. These metrics serve as intermediate representations that bridge the gap between measured signals and abnormality classification, enabling more accurate identification of target abnormalities without directly increasing the complexity of the measurement apparatus
2Reliability
If metrology measurements are performed on all targets, then comprehensive data is obtained, but time consumption and processing load increase
Solution Approach 1:
The patent applies local quality by focusing metrology measurement efforts on specific regions or targets that exhibit abnormal characteristics rather than uniformly measuring all targets. The target characterization module identifies and prioritizes abnormal targets based on metric analysis, allocating measurement resources locally to where they are most needed, thus maintaining data reliability while reducing overall measurement time
Solution Approach 2:
The patent implements partial action by performing metric calculation and target characterization selectively based on the needs of the specific measurement task. Rather than always performing complete analysis on all targets, the system can adjust the depth and scope of analysis, applying full characterization only when necessary to identify and resolve abnormalities, thereby reducing time consumption while maintaining necessary reliability
Data Source
AI summary
Methods and systems are provided, which identify specified metrology target abnormalities using selected metrics and classify the identified target abnormalities geometrically to link them to corresponding sources of error. Identification may be carried out by deriving target signals such as kernels from specified regions of interest (ROIs) from corresponding targets on a wafer, calculating the metrics from the target signals using respective functions and analyzing the metrics to characterize the targets.