Optical Image Contrast Metrics for Wafer Defect Target Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor manufacturing processes approach design rule limits, the detection of smaller defects becomes crucial for maintaining yield and electrical parameter integrity, but existing inspection systems struggle to differentiate between yield-relevant and nuisance defects, leading to increased complexity and cost in process corrections.
Innovation Solution
A system and method utilizing a light source, detector, and processor to generate an optical image, determine a gradient magnitude image, and calculate an image contrast metric by combining the contrast of specific pixel sets from the gradient magnitude and context range images, applying confidence levels to prioritize defect detection and alignment performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inspection sensitivity is increased to detect smaller defects, then defect detection capability is improved, but the number of nuisance defects detected increases dramatically
Solution Approach 1:
The patent changes the parameter of image contrast metric calculation by using gradient magnitude images and context range images with selective pixel sampling (less than 100% of pixels). This transforms the inspection approach to differentiate yield-relevant defects from nuisance defects through modified contrast metrics, resolving the contradiction between detection sensitivity and nuisance defect volume
Solution Approach 2:
The patent applies local quality by selecting specific pixels from gradient magnitude and context range images rather than processing all pixels uniformly. This selective sampling approach allows the system to focus computational resources on pixels most relevant to yield-affecting defects while filtering out nuisance defects, thus improving signal-to-noise ratio in defect detection
2Reliability
If all detected defects are corrected to eliminate defects, then yield is improved, but manufacturing complexity and cost increase dramatically
Solution Approach 1:
The patent extracts only the yield-relevant defects from the total defect population by using enhanced image contrast metrics and confidence level analysis. Instead of treating all detected defects equally, the system identifies and isolates the subset of defects that actually impact electrical parameters and yield, allowing focused process corrections on critical defects only
Solution Approach 2:
The patent creates a composite assessment by combining multiple image processing components (gradient magnitude image, context range image, selective pixel sampling, confidence levels) to generate a composite defect priority metric. This composite approach enables differentiation between yield-critical and non-critical defects, reducing unnecessary process complexity
3Measurement precision
If image processing is performed on all pixels, then measurement completeness is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies partial action by processing less than 100% of pixels in the image - specifically selecting pixels from gradient magnitude and context range images based on confidence levels. This selective pixel processing maintains sufficient measurement completeness for yield-relevant defect detection while significantly reducing computational load and processing time
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 enhances the sensitivity of defect detection, improves alignment quality, and reduces the complexity of identifying yield-relevant defects, thereby optimizing process control and reducing costs by focusing on systematic, electrically-relevant defects.
Implementation Method 1
a light source that generates a light beam; a stage configured to hold a wafer in a path of the light beam; a detector configured to receive the light beam reflected from the wafer
Data Source
AI summary
Global and local alignment energies are used in an image contrast metric. The image contrast metric can be used to find optical targets. Some pixels from a gradient magnitude image and a context range image from an optical image can be used to determine the image contrast metric. A heatmap from the image contrast metrics across part of a wafer can then be used to make a list of targets. Upper and lower confidence values can be applied to rank the available targets.