Wafer Surface Defect Inspection via Curve Chart Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing visual inspection methods for semiconductor wafers, particularly for determining flatness, are subjective and lack a consistent standard, leading to erroneous results due to reliance on human interpretation.
Innovation Solution
An automated wafer surface defect inspection method and apparatus that processes scanning information to identify defects by generating path information and analyzing parameters, such as radius and slice angle, to create curve charts, determining defect presence and type objectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated optical inspection (AOI) equipment is used to obtain haze images for wafer flatness determination, then inspection efficiency is improved, but subjective determination by visual inspectors still causes erroneous results
Solution Approach 1:
The patent replaces the mechanical/visual inspection system with an automated image processing system that uses computer algorithms to analyze haze images. The system automatically extracts features from the images and determines wafer flatness without human visual inspection, thereby eliminating subjective determination errors while maintaining high inspection efficiency.
Solution Approach 2:
The patent transforms the inspection process by changing from subjective visual parameter assessment to objective quantitative parameter measurement. The system extracts specific image parameters (such as haze distribution, intensity values, and spatial frequencies) and uses these measurable parameters to objectively determine wafer flatness and defects, replacing subjective visual judgment with precise parameter-based analysis.
2Adaptability or versatility
If manual visual inspection is performed by senior visual inspectors, then flexibility in judgment is maintained, but consistent standards are lacking leading to erroneous determination
Solution Approach 1:
The patent replaces the human visual inspection system with an automated computer-based image processing system. This substitution eliminates the inconsistency inherent in manual inspection while maintaining the ability to adapt to different inspection requirements through programmable algorithms and configurable parameters.
Solution Approach 2:
The patent creates a universal inspection system that can handle multiple types of wafer defects and inspection criteria through a single automated platform. The system is designed to be multi-functional, capable of performing various inspection tasks by adjusting processing parameters and algorithms, thereby providing both consistency and adaptability.
3Device complexity
If subjective determination by visual inspectors is used, then inspection process is simple, but excessive reliance on human eyes causes overly subjective inspection results
Solution Approach 1:
The patent replaces the simple but subjective human visual inspection process with an automated image processing system. While the system is more complex than manual inspection, it provides objective and precise results by using computer algorithms to analyze haze images, eliminating the subjectivity inherent in human judgment.
Solution Approach 2:
The patent uses digital copying of the wafer surface through haze imaging, followed by automated processing of these digital copies. The system creates digital representations of the wafer surface and performs all analysis on these copies, eliminating the need for direct human visual inspection while maintaining the ability to examine and analyze the wafer surface characteristics.
Data Source
AI summary
A wafer surface defect inspection method and a wafer surface defect inspection apparatus are provided. The method includes the following steps. Scanning information of a wafer is received, and the scanning information includes multiple scanning parameters. At least one reference point of the scanning information is determined, and path information is generated according to the at least one reference point and a reference value. Multiple first scanning parameters corresponding to the path information in the scanning parameters are obtained according to the path information to generate a curve chart. According to the curve chart, it is determined whether the wafer has a defect, and a defect type of the defect is determined.