Wafer Edge Bead Removal Inspection Using Marked Reference Segments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting incomplete edge bead removal on disk-like objects in semiconductor manufacturing are inefficient and unreliable, as they fail to effectively limit the inspection to specific marked areas, leading to unnecessary processing time and potential defects in subsequent layers.
Innovation Solution
A method that involves recording an image of a reference object with complete edge bead removal, marking structured elements on the reference object, and inspecting further objects by applying these marks to determine the completeness of edge bead removal, limiting the inspection to these marked areas and using a linear array camera with dark field illumination to identify photoresist residues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire peripheral area of the wafer is inspected, then the detection reliability is improved, but the inspection time increases
Solution Approach 1:
The peripheral area of the wafer is divided into multiple segments: a first peripheral area containing structured elements (dies) and a second peripheral area without structured elements. The inspection system selectively inspects only the first peripheral area, segmenting the inspection scope to eliminate redundant inspection of the second area while maintaining detection reliability for critical regions.
Solution Approach 2:
Different inspection strategies are applied to different regions of the wafer periphery. The first peripheral area with structured elements receives detailed inspection to ensure complete edge bead removal, while the second peripheral area is excluded from inspection. This local differentiation optimizes the balance between detection reliability and inspection time by focusing resources where defects would have the greatest impact.
2Productivity
If the inspection is limited to marked areas, then the processing time is reduced, but the detection reliability may be compromised
Solution Approach 1:
Marks are pre-defined on the wafer map to indicate the first peripheral area containing structured elements before inspection begins. This preliminary marking guides the inspection system to focus only on relevant areas, reducing processing time while ensuring that all critical regions with structured elements are inspected for edge bead removal completeness.
Solution Approach 2:
The inspection system uses optical detection methods with a linear array camera to detect edge bead removal status in the marked areas, replacing potential mechanical or manual inspection methods. This substitution enables rapid automated inspection of the first peripheral area while maintaining high detection reliability through optical precision.
3Productivity
If photoresist residues are not detected, then the processing continues, but the quality of subsequent layers deteriorates
Solution Approach 1:
The inspection system provides feedback by comparing the detected edge bead removal status in the first peripheral area against the pre-defined marks. When photoresist residues are detected, the system generates a signal indicating incomplete removal, allowing real-time process control decisions to be made to prevent defects in subsequent layers while maintaining overall process continuity for acceptable wafers.
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 allows for quick and reliable detection of incomplete edge bead removal by focusing the inspection on specific marked areas, reducing processing time and ensuring complete removal of photoresist residues, thereby improving the quality of subsequent layers.
Implementation Method 1
The peripheral area to be imaged is illuminated by a light source in a dark field array
Data Source
AI summary
A method of detecting incomplete edge bead removal from a disk-like object is disclosed. First a peripheral area of a disk-like reference object is imaged. Marks are then defined in the peripheral area of the reference object. Finally, images of peripheral areas of a plurality of disk-like objects of the same batch are recorded. The inspection of the disk-like objects is limited to the locations of the marks defined on the reference object.


