Inspection Method Using Polynomial Autofocus for Semiconductor Masks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pattern inspection methods for semiconductor masks lack precision in detecting ultra-fine defects, particularly due to limitations in focusing control when using reflected light, which is weak and pattern-dependent, affecting the accuracy of defect detection and yield improvement in semiconductor manufacturing.
Innovation Solution
The method involves acquiring a first image using an auto focuser, calculating a two-dimensional polynomial approximation of autofocus function coordinates, and controlling focusing for a second image acquisition, allowing for improved focus control even with weak reflected light and reducing pattern dependency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If an auto focuser using reflected light is used to acquire images for pattern inspection, then the inspection system can be configured, but the focusing control precision deteriorates due to weak and pattern-dependent reflected light
Solution Approach 1:
The patent introduces a two-dimensional polynomial approximation as an intermediary model between the raw autofocus function coordinates and the focusing control. This mathematical model smooths out the pattern-dependent variations in reflected light by fitting a polynomial surface to the measured coordinates, providing a more stable and precise focusing control that is less sensitive to light intensity fluctuations
Solution Approach 2:
The patent transforms the focusing control from directly using raw autofocus function coordinates to using a two-dimensional polynomial approximation of these coordinates. This parameter transformation changes the control variables from noisy, pattern-dependent measurements to smoothed, model-based parameters that provide more consistent focusing precision across different pattern types
2Ease of manufacture
If reflected light is used for autofocus in pattern inspection, then the system can operate with the given optical setup, but the detection precision of ultra-fine defects deteriorates due to pattern dependency
Solution Approach 1:
The patent implements a feedback mechanism where the two-dimensional polynomial approximation of autofocus function coordinates is used to continuously adjust and optimize focusing control. This feedback loop compensates for pattern-dependent variations by referencing the mathematical model, ensuring consistent focusing precision regardless of the specific pattern being inspected, thereby improving defect detection accuracy
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 precision of defect inspection by enabling effective focus control with reduced pattern dependency, thereby improving the detection of ultra-fine defects and increasing the yield in semiconductor manufacturing.
Implementation Method 1
acquiring a first image of the sample to be inspected, with an auto focuser using the illuminating light reflected from the sample to be inspected
Data Source
AI summary
Provided is an inspection method including: irradiating a sample to be inspected with illuminating light; acquiring a first image of the sample to be inspected, with an auto focuser using the illuminating light reflected from the sample to be inspected; storing autofocus function coordinates acquired with the auto focuser in the acquiring the first image; calculating a two-dimensional polynomial approximation of the autofocus function coordinates; and controlling focusing in acquiring a second image of the sample to be inspected, with the two-dimensional polynomial approximation.