Wafer Inspection Defect Sizing Using Image Sensor Charge Overflow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wafer inspection systems face challenges in accurately determining defect sizes across a large measurement range due to the dramatic change in light scattering, which is proportional to the sixth power of defect size, making it difficult to achieve precise measurements.
Innovation Solution
The system employs an image sensor without an anti-blooming feature, allowing excess charge to flow to neighboring pixels when a pixel reaches full well capacity, and a computer subsystem to determine defect sizes using the output from both the pixel and its neighboring pixels, thereby extending the defect sizing range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an image sensor with anti-blooming feature is used, then the pixel can maintain full well capacity, but the defect sizing range is limited
Solution Approach 1:
The patent converts the harmful effect of charge overflow (blooming) into a beneficial measurement mechanism. By removing the anti-blooming feature, excess charge naturally flows to neighboring pixels, and the system uses this overflow pattern to extend the measurement range. The computer subsystem analyzes the charge distribution across multiple pixels to determine defect sizes beyond the single-pixel full well capacity limit.
Solution Approach 2:
The patent transitions from single-pixel measurement to multi-pixel measurement by utilizing the spatial dimension of charge overflow. Instead of relying on a single pixel's full well capacity, the system distributes the measurement across multiple neighboring pixels, effectively extending the dynamic range by utilizing the spatial arrangement of pixels rather than increasing individual pixel capacity.
2Measurement precision
If light scattering detection is used, then defect size can be determined from scattered light quantity, but accurate measurement across large range is difficult due to sixth power proportionality
Solution Approach 1:
The patent segments the measurement task across multiple pixels instead of relying on a single pixel to capture the entire light scattering signal. By dividing the detection function across multiple pixels, the system can handle a broader range of scattering intensities without saturation, effectively extending the measurable defect size range while maintaining accuracy through distributed measurement.
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 enables more accurate and extended defect sizing range, potentially increasing it by 10,000 times, allowing for more precise defect size determination across a broader range compared to traditional systems with anti-blooming features.
Implementation Method 1
Some inspection systems are configured to detect defects by detecting the light scattered from wafer defects. In general, defect size can be determined from the quantity of the light scattered from a wafer defect.
Implementation Method 2
an image sensor configured to detect light scattered from wafer defects and to generate output responsive to the scattered light
Data Source
AI summary
Various embodiments for extended defect sizing range for wafer inspection are provided. One inspection system includes an illumination subsystem configured to direct light to the wafer. The system also includes an image sensor configured to detect light scattered from wafer defects and to generate output responsive to the scattered light. The image sensor is also configured to not have an anti-blooming feature such that when a pixel in the image sensor reaches full well capacity, excess charge flows from the pixel to one or more neighboring pixels in the image sensor. The system further includes a computer subsystem configured to detect the defects on the wafer using the output and to determine a size of the defects on the wafer using the output generated by a pixel and any neighboring pixels of the pixel to which the excess charge flows.