Dark-Field Imaging for High-Aspect Ratio Groove Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optical inspection techniques are inadequate for detecting defects in high-aspect ratio grooves, such as vias in semiconductor wafers, due to their narrow and deep geometry, which prevents effective identification of buried defects using standard imaging methods.
Innovation Solution
The implementation of a dark-field imaging technique that utilizes a large field of view to analyze pixel brightness, differentiating between normal and defective vias through their scattering and absorption properties, allowing for the identification and classification of defective features in patterned structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard optical imaging techniques are used, then the inspection system is simple and fast, but it cannot detect defects in high-aspect ratio grooves due to their narrow and deep geometry
Solution Approach 1:
The patent changes the illumination parameters by using oblique incident light at specific angles (e.g., 45 degrees) instead of normal incidence, and adjusts the detection angle to collect scattered light. This parameter change enables the visualization of high-aspect ratio groove defects that are invisible to standard normal-incidence optical imaging, resolving the contradiction between detection capability and system complexity.
Solution Approach 2:
The patent transitions from standard 2D optical imaging to a configuration that effectively utilizes angular dimensions by positioning the light source and detector at specific angles relative to the sample. This angular dimensionality change allows light to penetrate and scatter within the deep grooves, making buried defects detectable without significantly increasing system complexity.
2Productivity
If the field of view is increased to cover multiple grooves, then the inspection speed increases, but the resolution for detecting small defects in individual grooves decreases
Solution Approach 1:
The patent applies local quality by using oblique illumination that creates enhanced light scattering specifically within the groove regions while maintaining a larger field of view. The angled light interacts differently with groove structures compared to flat surfaces, providing localized contrast enhancement that preserves defect detection precision across multiple grooves simultaneously, thus resolving the contradiction between inspection speed and precision.
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 method enables fast and reliable detection of defects within high-aspect ratio grooves, providing effective on-line inspection and sorting capabilities, ensuring the quality of semiconductor wafers by identifying defects that standard imaging techniques cannot detect.
Implementation Method 1
The implementation of a dark-field imaging technique that utilizes a large field of view to analyze pixel brightness, differentiating between normal and defective vias through their scattering and absorption properties
Implementation Method 2
The implementation of a dark-field imaging technique that utilizes a large field of view to analyze pixel brightness, differentiating between normal and defective vias through their scattering and absorption properties
Data Source
AI summary
An inspection system and method are presented for inspecting structures having a pattern formed by an array of elongated grooves having high aspect-ratio geometry, such as semiconductor wafers formed with vias. The inspection system comprises an imaging system and a control unit. The imaging system is configured and operable for imaging the structure with a dark-field imaging scheme and generating a dark-field image. The control unit comprises an analyzer module for analyzing pixels brightness in the dark-field image for identifying a defective groove, being a groove characterized by pixels brightness in the dark-field image lower than nominal brightness by a predetermined factor.