Wafer Defect Sizing via Multi-Channel Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining defect sizes on semiconductor wafers, such as using defect pixel count or scattering intensity, are inaccurate due to factors like defect shape, material, and surface roughness, particularly for patterned wafers, leading to misleading size reports and poor defect detection.
Innovation Solution
A computer-implemented method that separates defects into groups and subgroups based on output from multiple channels of an inspection system, using calibration parameters specific to each subgroup, acquired from actual defect measurements on other wafers, to accurately determine defect sizes by combining scattering intensity and area information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If defect pixel count is used to report defect size, then the measurement process is simple, but the measurement precision is poor due to factors like defect shape, material, and surface roughness
Solution Approach 1:
The patent segments the defect analysis process into multiple channels (e.g., dark field, bright field, scatter) that independently measure different aspects of defect characteristics. Each channel provides specific information about defect size, shape, and material properties, which are then combined to achieve accurate defect sizing that accounts for various influencing factors without requiring complex single-measurement processes
Solution Approach 2:
The patent changes the measurement parameters by using multiple detection channels with different physical principles (scattering intensity, absorption, reflection) instead of relying on a single parameter like pixel count. This allows the system to adapt measurements to different defect types, materials, and surface conditions, thereby improving measurement precision while maintaining processability through automated multi-parameter analysis
2Ease of operation
If scattering intensity from a single channel is used to report defect size, then the measurement process is straightforward, but the measurement precision is insufficient for patterned wafers and various defect types
Solution Approach 1:
The patent divides the inspection system into multiple operational channels (dark field, bright field, scatter channels) that can be independently operated or combined. Each channel is optimized for detecting specific defect characteristics, and the system selectively activates appropriate channels based on defect type and wafer pattern complexity, maintaining operational simplicity while improving measurement precision through targeted multi-channel analysis
Solution Approach 2:
The patent creates a universal defect measurement system that can handle various defect types (particles, voids, bridges) and wafer patterns (patterned, unpatterned) through a single integrated multi-channel platform. The system automatically selects and combines appropriate measurement channels based on the inspection requirements, providing accurate defect sizing across diverse applications without requiring separate specialized systems
3Measurement precision
If multiple channels are used to separate defects into groups and subgroups with calibration parameters, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent segments the complex multi-channel system into modular functional units, each handling specific defect characteristics. The system divides defect analysis into hierarchical groups and subgroups based on initial channel measurements, then applies calibration parameters selectively to relevant subgroups. This modular segmentation reduces overall system complexity by organizing multiple channels into manageable, independently configurable modules while maintaining high measurement precision through coordinated operation
Solution Approach 2:
The patent applies local quality by using different calibration parameters and analysis methods for different defect groups and subgroups rather than applying a uniform complex algorithm to all defects. Each defect category receives tailored calibration and measurement approaches based on its specific characteristics, reducing the need for overly complex universal algorithms while achieving high precision for each defect type through localized optimization
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 accuracy of defect size reporting, allowing for more precise monitoring and control of semiconductor processes by categorizing defects effectively and using ground-truth data for calibration, improving defect size determination across various types and sizes.
Implementation Method 1
defects detected on a wafer... output acquired for the defects by multiple channels of an inspection system... light scattered from patterns or other features on the specimen will be combined with other scattered light
Data Source
AI summary
Computer-implemented methods, carrier media, and systems for determining sizes of defects detected on a wafer are provided. One computer-implemented method includes separating the defects into groups based on output acquired for the defects by multiple channels of an inspection system used to detect the defects on the wafer. The method also includes separating the defects in one or more of the groups into subgroups based on the output acquired for the defects by one or more of the multiple channels. In addition, the method includes determining the sizes of one or more of the defects in one or more of the subgroups separately based on the output acquired for the defects by only one of the multiple channels and a calibration parameter. The calibration parameter is different for each of the subgroups and is acquired by using another system to measure actual sizes of defects detected on other wafers.