Pattern Defect Detection via Time-Resolved Electrical Signal Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor processes become miniaturized and complex, detecting fine pattern defects becomes increasingly challenging due to the reduced size of defects and the need for higher inspection sensitivity, making it difficult to accurately identify electrical characteristics changes caused by these defects.
Innovation Solution
A method involving the application of voltage to an object to be inspected, measuring inspection signals over time, generating intensity images showing the relationship between signal intensity and time, and comparing these images with comparative images to detect pattern defects by analyzing changing trends rather than just intensity differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If voltage is applied and inspection signal is measured over time to generate intensity images, then detection efficiency and accuracy of pattern defects is improved, but inspection time and measurement complexity increases
Solution Approach 1:
The patent applies voltage to the object to be inspected before the actual inspection process, allowing the object to reach a stable electrical state. This preliminary voltage application ensures that subsequent inspection signals reflect only the defects being sought, not transient electrical instabilities, thereby improving measurement precision without requiring repeated measurements that would increase inspection time
Solution Approach 2:
The patent measures inspection signals over time while voltage is applied, capturing dynamic changes in electrical characteristics. By analyzing the temporal evolution of inspection signals rather than static values, the system can distinguish between normal variations and actual defects, improving detection accuracy while maintaining efficient inspection throughput through intelligent signal processing
2Reliability
If inspection signal is measured over time and intensity images are generated showing relationship between signal intensity and time, then pattern defects can be detected that would be missed by static comparisons, but device complexity and processing requirements increase
Solution Approach 1:
The patent transforms the inspection data from a simple spatial distribution into a temporal-spatial intensity image that shows the relationship between signal intensity and time. This additional temporal dimension allows defects to be identified through their characteristic time-evolution patterns, significantly improving detection reliability while the image processing algorithms manage the complexity through efficient data representation
Solution Approach 2:
The patent introduces intensity images as an intermediary representation between the raw inspection signals and the final defect detection. These images consolidate temporal signal variations into visual patterns that are easier to analyze and compare, reducing the complexity of the detection algorithm while maintaining high reliability in defect identification
3Measurement precision
If voltage is applied to object to be inspected, then inspection signal generated in pattern can be measured, but electrical characteristics changes caused by fine pattern defects become more difficult to detect due to miniaturization
Solution Approach 1:
The patent changes the measurement parameter from static physical dimensions to dynamic electrical characteristics. By applying voltage and measuring inspection signals that reflect electrical resistance changes over time, the system can detect fine pattern defects through their electrical impact rather than their physical size, overcoming the limitations of miniaturization and improving measurement precision for sub-micron 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 improves the detection efficiency of pattern defects by analyzing signal changes over time, allowing for the identification of defects that would be missed by static intensity comparisons, thereby enhancing the accuracy and reliability of semiconductor manufacturing processes.
Implementation Method 1
measuring an inspection signal that varies according to an electrical resistance of a pattern of an object to be inspected
Data Source
AI summary
Provided is a method of inspecting a pattern defect. The method includes: applying a voltage to an object to be inspected and measuring an inspection signal generated in a pattern of the object to be inspected due to the voltage applied to the object to be inspected over time; generating an intensity image showing a relationship between an intensity of the inspection signal measured in the pattern and a time by processing the inspection signal; and detecting a pattern defect position by comparing the intensity image with a comparative intensity image.