Charged Particle Defect Assessment with Error-Aware Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing charged particle assessment systems face challenges in distinguishing noise from genuine defects, leading to excessive rejection of substrates and reduced yield in semiconductor manufacturing.
Innovation Solution
The proposed solution involves an assessment method and system that generate assessment signals representing a sample's surface properties, process these signals to identify candidate defects, and monitor for error conditions. The system analyzes candidate defect signals to determine if defects are real, and if an error condition is detected, it stops the analysis of the candidate defect to prevent false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If statistical means and thresholds are used to eliminate noise, then some noise can be eliminated, but genuine defects cannot be distinguished from all noise such as nuisance signals
Solution Approach 1:
A status signal acts as an intermediary between the assessment signals and the defect analysis process. This status signal indicates whether error conditions occurred during assessment signal generation, allowing the system to filter out data affected by errors before definitive defect classification, thereby reducing false positives while maintaining detection accuracy
Solution Approach 2:
The system performs preliminary monitoring of error conditions during the assessment signal generation phase. By detecting error conditions early and generating status signals before defect analysis, the system can preemptively identify and exclude problematic data, preventing false defect identification rather than attempting to correct errors after analysis
2Measurement precision
If all candidate defects are analyzed to ensure no genuine defects are missed, then detection completeness is improved, but processing time increases and throughput decreases
Solution Approach 1:
The system extracts and removes assessment signals affected by error conditions from the analysis pipeline using status signals. By taking out problematic data early in the process, the system avoids wasting processing resources on data that would produce false results, thereby maintaining detection completeness for valid data while improving overall processing efficiency and throughput
3Measurement precision
If noise reduction techniques are applied to improve signal-to-noise ratio, then image quality is improved, but genuine defects may be indistinguishable from remaining noise
Solution Approach 1:
The status signal serves as an intermediary that provides metadata about the quality and reliability of assessment signals. Rather than relying solely on signal-to-noise ratio improvements, this intermediary information allows the system to identify and exclude compromised data, preserving defect signal integrity without requiring aggressive noise reduction that might mask genuine defects
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively reduces false positives by identifying and excluding data affected by error conditions, thereby improving the accuracy of defect detection and maintaining high substrate throughput in semiconductor manufacturing.
Implementation Method 1
The interactions between the material structure at the probing spot and the landing electrons from the beam of electrons cause signal electrons to be emitted from the surface, such as secondary electrons, backscattered electrons or Auger electrons
Data Source
AI summary
An assessment method comprising: using an assessment apparatus to generate assessment signals representing a property of a surface of a sample; processing the assessment signals to identify candidate defects and outputting a candidate defect signal; monitoring the status of the assessment apparatus for error conditions and generating a status signal indicating any error conditions during functioning of the assessment apparatus; and analysing the candidate defect signal to determine if the candidate defects are real defects; wherein analysis of a candidate defect is not completed if the status signal indicates that the assessment signal(s) and/or the candidate defect signal corresponding to the candidate defect would have been affected by an error condition.