SEM Line Width Measurement Quality Decision via Edge Taper Width Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automatic measurement of line width and other pattern dimensions in SEMs often fails due to variance in measurement values, autofocusing issues, and image drift caused by charging, making it difficult to determine the correctness of measurements without human intervention.
Innovation Solution
The method decides the quality of a measurement value based on the taper widths of signal intensity distributions at the edge parts of the pattern, determining a measurement to be correct if the taper widths fall within a predetermined range, allowing for automatic detection of defective measurements attributed to focusing or charging issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic measurement is performed using SEM, then productivity is improved, but measurement precision deteriorates due to variance and failures
Solution Approach 1:
The invention introduces a feedback mechanism where the measurement quality is evaluated based on the standard deviation of multiple measurement values. When the standard deviation exceeds a threshold, the system automatically requests re-measurement, creating a closed-loop quality control system that ensures measurement precision while maintaining automation.
Solution Approach 2:
The system performs self-validation by automatically assessing the quality of its own measurements through statistical analysis. The measurement apparatus evaluates its own output quality and triggers re-measurement when quality criteria are not met, eliminating the need for external human judgment while maintaining high precision.
2Measurement precision
If measurement quality is judged by human observation, then measurement precision is improved, but device complexity increases and automation is lost
Solution Approach 1:
The invention replaces the mechanical/human observation system with an automated computational system. Instead of human operators visually assessing measurement quality, the system uses automatic statistical calculations (standard deviation computation) and algorithmic decision-making to evaluate and judge measurement quality, thereby maintaining precision while achieving full automation.
Solution Approach 2:
The invention transforms the quality judgment process from a subjective visual assessment to an objective parameter-based evaluation. By converting quality judgment into a quantitative parameter (standard deviation threshold comparison), the system achieves both automation and consistent precision without human intervention.
3Ease of operation
If simple range checking is used for quality decision, then ease of operation is improved, but measurement precision deteriorates due to inability to handle variance
Solution Approach 1:
The invention enhances the simple range-checking approach by introducing a statistical parameter (standard deviation) as the quality criterion. Instead of using fixed predetermined ranges that cannot adapt to variance, the system dynamically evaluates measurement quality based on the calculated standard deviation of multiple measurements, maintaining operational simplicity while accurately handling measurement variance.
Data Source
AI summary
A method of deciding the quality of a measurement value of the line width, the line interval or the like of a pattern on an object to-be-measured, including acquiring the signal intensity distribution of the pattern on the object to-be-measured, detecting the edge positions of the pattern from the acquired signal intensity distribution, detecting the taper widths of the edge parts of the pattern from the acquired signal intensity distribution, and deciding that the measurement value calculated on the basis of the detected edge positions is correct, when the detected taper widths fall within a predetermined range set beforehand. In this way, it is permitted to automatically decide the defective measurement of the line width of the pattern, or the like, attributed to an unclear image due to inferior focusing in an image photographing mode, an unclear image due to an image drift ascribable to charging-up, or the like.