Overlay Metrology Target Centering Using ML Stage Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional metrology tools for overlay measurement are slow and require well-focused acquisition images, making them inefficient and inaccurate for precise alignment of features on multiple sample layers.
Innovation Solution
A system and method utilizing a machine learning-based centering model to determine stage correctables, adjust stage positions, and acquire measurement images, thereby improving focus and alignment efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional hardware-based focusing mechanisms and image processing algorithms are used for target decentering detection, then measurement precision can be maintained, but the measurement speed becomes slow and requires well-focused acquisition images
Solution Approach 1:
The patent replaces conventional hardware-based focusing mechanisms (mechanical/optical systems) with a machine learning-based centering model that processes images computationally. This substitution enables faster processing by using software-based autofocus and decentering detection algorithms instead of physical focusing mechanisms, directly resolving the contradiction between maintaining measurement precision and improving measurement speed.
Solution Approach 2:
The patent changes the operational parameters by using machine learning models that can process images across a range of focus positions rather than requiring a single well-focused image. The system acquires images at multiple focus positions and uses the ML model to determine the best focus position and decentering, allowing measurements to be performed faster by not requiring perfect initial focusing conditions.
2Measurement precision
If conventional methods require well-focused acquisition images for accurate decentering detection, then measurement precision is maintained, but the system becomes inefficient and time-consuming
Solution Approach 1:
The patent applies preliminary action by acquiring images at multiple focus positions before final measurement. The machine learning model analyzes these pre-acquired images to determine the optimal focus position and decentering corrections, allowing the system to prepare and correct stage positions in advance before acquiring the final measurement image, thereby reducing total measurement time while maintaining precision.
Solution Approach 2:
The patent replaces the conventional approach that requires manual or mechanical adjustment to achieve perfect focus with an automated machine learning-based autofocus system. The ML model rapidly analyzes images to determine focus quality and decentering, eliminating time-consuming manual focusing steps and mechanical adjustments, thus reducing loss of time while maintaining target centering accuracy.
3Reliability
If hardware-based focusing mechanisms are used, then focus control can be achieved, but the device complexity increases and requires additional hardware components
Solution Approach 1:
The patent replaces complex hardware-based focusing mechanisms with a computational approach using machine learning models. Instead of additional physical focusing components, the system uses software-based analysis of images at multiple focus positions to determine optimal focus and decentering corrections, thereby maintaining focus control reliability while reducing device complexity by eliminating the need for specialized hardware focusing mechanisms.
Data Source
AI summary
A system for target centering detection may be configured receive one or more acquisition images of a sample from an overlay metrology sub-system and determine, using a machine learning-based centering model, one or more stage correctables based on the received one or more acquisition images. The system may be configured to cause a sample stage of the overlay metrology sub-system to adjust a stage position based on the determined one or more stage correctables and receive one or more measurement images of the sample from the overlay metrology sub-system based on the adjusted stage position of the sample stage. The system may then be configured to determine one or more overlay measurements based on the received one or more measurement images.