Dynamic Digital Filtering for Semiconductor Wafer Surface Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surface inspection methods for semiconductor wafers face challenges in detecting foreign particles and defects due to low frequency fluctuations in reflected light, which are influenced by surface conditions, film type, thickness, and movement vibrations, leading to reduced sensitivity and accuracy.
Innovation Solution
The method involves dynamic digital filtering of inspection signals, where parameters such as cut-off frequency are varied based on the inspection object's movement speed, coordinate position, illumination spot size, film type, and surface characteristics to remove undesired low frequency components, allowing for accurate detection without raising the detection threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If analogue filter is used to remove low frequency component, then low frequency fluctuation is reduced, but detection sensitivity deteriorates due to raised threshold level
Solution Approach 1:
The patent replaces the analogue filter with a digital filter that operates after A/D conversion. The digital filter processes the signal in the digital domain, allowing for more precise control of the filtering process without the limitations of analogue circuits. This substitution enables flexible adjustment of filter parameters and eliminates the need to raise the threshold level, thereby maintaining detection sensitivity while removing low frequency fluctuations.
Solution Approach 2:
The patent dynamically adjusts the cut-off frequency of the digital filter based on inspection conditions such as wafer rotation speed, illumination spot size, and surface characteristics. By changing the filter parameter (cut-off frequency) according to actual inspection conditions, the system optimizes the removal of low frequency components while preserving signal integrity and detection sensitivity for foreign particles of various sizes.
2Ease of operation
If fixed threshold is used for detection, then simplicity is maintained, but detection accuracy decreases due to varying surface conditions and film properties
Solution Approach 1:
The patent introduces dynamic threshold adjustment based on the extracted low frequency component characteristics. Instead of using a fixed threshold, the system calculates an adaptive threshold that varies with inspection conditions including wafer rotation speed, illumination parameters, and surface properties. This dynamic thresholding enables accurate detection across different surface conditions and film types while maintaining operational simplicity through automated adjustment.
Solution Approach 2:
The patent implements a feedback mechanism where the extracted low frequency component is used to dynamically adjust the detection threshold. The system continuously monitors the signal characteristics and adjusts the threshold accordingly, creating a closed-loop control system. This feedback approach ensures optimal detection accuracy for foreign particles while adapting to varying surface conditions and film properties in real-time.
3Object-affected harmful factors
If wide attenuation frequency band is used in filter, then low frequency component removal is improved, but signal distortion increases due to margin requirement
Solution Approach 1:
The patent dynamically adjusts the cut-off frequency parameter of the digital filter based on specific inspection conditions such as wafer rotation speed, illumination spot size, and surface characteristics. By optimizing the cut-off frequency for each inspection scenario, the system achieves effective low frequency component removal while minimizing signal distortion. This parameter adaptation allows the attenuation band to be wide enough to capture low frequency components without excessively broadening to distort useful signal information.
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 accurate detection of foreign particles and defects on wafers with varying surface conditions, maintaining high sensitivity and throughput by reducing low frequency noise, allowing for precise measurement and consistent performance across different wafer types.
Implementation Method 1
a scattered light on a wafer surface by a light receiving element like a photoelectron multiplier tube and analyze the extent of light scattering
Implementation Method 2
a method to detect a scattered light on a wafer surface by a light receiving element like a photoelectron multiplier tube
Data Source
AI summary
A semiconductor wafer, which is an inspection object, is stuck by vacuum on a chuck and this chuck is mounted on an inspection object movement stage consisting of a rotational stage and a translational stage, located on a Z-stage. The rotational stage provides a rotational movement and the translational stage provides a translational movement. And when a foreign particle or a defect on an inspection object surface is detected, the parameter of digital filtering is dynamically changed during inspection, and the foreign particle or the defect is differentiated using the result after removing a low frequency fluctuation component to be a noise component.