Wafer Defect Inspection Using Exposure Pattern Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional wafer defect inspection methods are inefficient and inaccurate as they rely on pre-defined killer defects, failing to identify unknown defects and requiring extensive manual analysis by engineers.
Innovation Solution
A wafer defect inspection system that uses a defect inspection apparatus to obtain a defect distribution map and compares it with a first exposure pattern to divide defects into killer and non-killer groups based on their locations, allowing engineers to focus analysis only on critical defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional defect inspection method uses pre-defined killer defects for inspection, then known killer defects can be determined, but unknown defects cannot be identified and inspection accuracy is insufficient
Solution Approach 1:
The system performs preliminary classification of defects into killer and non-killer groups based on their spatial relationship with component patterns before detailed analysis. By pre-establishing the exposure pattern and comparing defect locations against it, the system prepares classification criteria in advance, enabling accurate identification of both known and unknown defect types without requiring pre-definition of all possible defect categories.
2Measurement precision
If all defects are analyzed manually by engineers, then comprehensive defect analysis can be performed, but inspection efficiency is low and workload is high
Solution Approach 1:
The system extracts and separates killer defects from non-killer defects based on their locations relative to component patterns. By automatically classifying defects into two distinct groups, the system extracts only the critical killer defects that require detailed engineering analysis, while non-killer defects are automatically handled. This significantly reduces the workload for engineers while maintaining comprehensive analysis of critical issues.
Solution Approach 2:
The system performs self-service by automatically classifying defects into killer and non-killer groups using the exposure pattern comparison method. This automated classification reduces dependence on manual engineering analysis for routine defect sorting, allowing engineers to focus only on complex killer defects that require human judgment, thereby improving overall inspection efficiency.
3Productivity
If defect distribution map is compared with exposure pattern to divide defects into groups, then engineer workload is reduced and inspection efficiency is enhanced, but system complexity increases
Solution Approach 1:
The system uses a copy of the exposure pattern (the design template) and overlays it with the defect distribution map. By comparing defect locations against this copied reference pattern, the system automatically determines whether defects fall in killer or non-killer regions. This copying approach simplifies the comparison process while maintaining accuracy, as it avoids complex geometric calculations and directly leverages the existing exposure pattern data.
Data Source
AI summary
A defect inspection method for a wafer is provided. The wafer comprises a component pattern. The method comprises the following steps: providing a defect inspection apparatus for inspecting the defects on the wafer to obtain a defect distribution map; providing a photo mask, wherein the photo mask comprises a exposure pattern corresponding to the component pattern; and comparing the defect distribution map with the exposure pattern and dividing the defects in the defect distribution map into a first killer defect group and a first non-killer defect group according to their corresponding locations in the exposure pattern. In addition, a wafer defect inspection system applying the same method is also provided.