Wafer Edge Defect Detection via Virtual Image Stacking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting defects on the edges of wafers, particularly semiconductor wafers, is challenging due to their brittle nature, which can lead to cracks and breakage, and existing inspection methods lack accuracy without sacrificing throughput.
Innovation Solution
A method and system that captures images of wafer edges using a metrology unit and combines them into a virtual stack to identify defects, enhancing detection accuracy by analyzing patterns across multiple images without physically stacking wafers, thus improving defect identification and detection without reducing production speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical stacking of wafers is used to inspect edge defects, then defect detection accuracy is improved, but production throughput is reduced due to manual handling time
Solution Approach 1:
The patent creates virtual copies of wafer edge images through computational imaging, combining multiple 2D images to form a 3D virtual stack that replicates the defect detection capabilities of physical stacking without the manual handling requirements. This allows automated inspection maintaining both high accuracy and throughput.
Solution Approach 2:
The patent replaces the mechanical physical stacking process with a computational imaging system that uses software algorithms to combine images. This substitution eliminates manual wafer stacking while achieving the same defect detection accuracy through virtual image reconstruction.
2Measurement precision
If automated image combining is used to form virtual stacks, then defect identification accuracy is improved, but system complexity increases
Solution Approach 1:
The patent implements a multi-functional inspection system where a single automated imaging platform performs multiple functions: capturing wafer edge images, combining images computationally, forming virtual stacks, and identifying defects. This integration reduces overall system complexity compared to separate specialized systems.
Data Source
AI summary
Embodiments described herein provide for a defect detection system and method suitable for detecting defects on an edge of a wafer. The method includes placing at least two wafers sequentially on a conveyor. Images of at least the edges of each wafer placed on the conveyor are captured and sent to a controller. A defect detection software combines the images to show the edges of the wafers in a virtual stack. The virtual stack allows for a pattern of defects to be identified. The pattern of defects in close proximity will allow for identification of the defects in the edges of the wafers.