Pupil Plane Calibration for Scatterometry Overlay Metrology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Miscalibration of the pupil plane in scatterometry overlay metrology leads to errors and degrades metrology performance, particularly due to the limitations of existing methods that rely on electromagnetic modeling and mirror-like calibration targets.
Innovation Solution
A method for calibrating the pupil center by calculating fluctuations in the overlay signal per pixel and minimizing these fluctuations with respect to a pupil weighted variance, which does not require electromagnetic modeling and is applicable to various scatterometry overlay targets with overlapping or non-overlapping cells.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If electromagnetic modeling and mirror-like calibration targets are used to calibrate the pupil plane, then calibration can be performed, but measurement precision deteriorates due to tool-induced shift and its variance
Solution Approach 1:
The patent uses a virtual copy of the calibration target in software rather than a physical mirror-like calibration target. The virtual calibration target is generated computationally and used to calculate the point spread function, eliminating the need for specialized physical calibration samples while achieving accurate pupil plane calibration and reducing tool-induced shift.
Solution Approach 2:
The patent replaces the physical electromagnetic modeling approach with a computational software-based approach. Instead of using Maxwell equations and electromagnetic simulations with mirror-like targets, the method uses image processing algorithms to analyze calibration target images and compute the point spread function, substituting mechanical/physical calibration systems with software-based processing.
2Measurement precision
If electromagnetic modeling is used for pupil plane calibration, then calibration can be achieved, but device complexity increases due to the need for complex modeling and specialized calibration targets
Solution Approach 1:
The patent creates a virtual calibration target through software computation rather than requiring a physical mirror-like calibration target. This virtual copy is generated from the calibration target image and used to compute the point spread function, eliminating the need for specialized physical calibration hardware while maintaining calibration accuracy.
Solution Approach 2:
The patent extracts and removes the complex electromagnetic modeling step from the calibration process. By taking out the Maxwell equations and electromagnetic simulation requirements, the method simplifies the calibration system to use only image processing and computational algorithms, reducing device complexity while preserving measurement precision.
3Measurement precision
If conventional calibration methods are used, then pupil plane calibration can be performed, but productivity decreases due to the time-consuming nature of electromagnetic modeling
Solution Approach 1:
The patent replaces time-consuming electromagnetic modeling computations with faster image processing algorithms. The software-based method processes calibration target images directly to compute the point spread function, eliminating the need for iterative Maxwell equations solutions and significantly reducing calibration time while maintaining accuracy.
Solution Approach 2:
The patent uses a computationally generated virtual calibration target that can be processed rapidly through image processing algorithms. This virtual copy allows for fast calculation of the point spread function without requiring repeated electromagnetic simulations, thereby improving calibration speed and productivity while preserving measurement precision.
Data Source
AI summary
Methods and calibrations modules are provided, for calibrating a pupil center in scatterometry overlay measurements. The calibration comprises calculating fluctuations from a first statistical figure of merit such as an average of an overlay signal per pixel at the pupil and significantly reducing, for example minimizing, the fluctuations with respect to a second statistical figure of merit thereof, such as a pupil weighted variance of the fluctuations.