Topography Determination via Computational Lithography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lithography apparatuses struggle to measure and correct product-induced topography at sub-millimeter lateral resolution, which is significant compared to process focus margins, making it difficult to identify and rank hotspots prone to defects due to topography.
Innovation Solution
A method combining computational lithography modeling and on-product measurement to determine topography by obtaining focus values from both unpatterned and patterned substrates, allowing for accurate identification and ranking of hotspot criticality based on topography data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional level sensors are used to measure substrate topography, then measurement is possible, but the lateral resolution is insufficient to capture sub-millimeter product-induced topography
Solution Approach 1:
The patent replaces the mechanical level sensor measurement system with a computational approach using lithography exposure data. By analyzing focus variations in measured aerial images through computational modeling, the system achieves sub-millimeter topography measurement capability that conventional mechanical sensors cannot provide
Solution Approach 2:
The patent introduces computational lithography modeling and aerial image measurement as an intermediary between the substrate topography and the measurement system. Instead of directly measuring topography with insufficient-resolution sensors, the system measures aerial images and uses computational models to extract topography information at the required sub-millimeter resolution
2Manufacturing precision
If focus control systems use conventional level sensor data, then focus adjustment is possible, but correction of sub-millimeter topography is ineffective due to physical limitations of exposure slits
Solution Approach 1:
The patent implements a feedback loop where measured aerial images are compared against computational lithography models to determine actual substrate topography. This feedback information is then used to adjust focus control, enabling effective correction of sub-millimeter topography that directly improves pattern fidelity
Solution Approach 2:
The patent changes the approach from direct mechanical focus adjustment based on level sensor data to computational focus optimization based on aerial image analysis. By changing the measurement parameters and using computational modeling, the system achieves effective topography correction without requiring complex hardware modifications
3Measurement precision
If computational lithography modeling alone is used to identify hotspots, then prediction is possible, but accuracy is insufficient without actual product topography data
Solution Approach 1:
The patent performs preliminary computational lithography modeling to predict potential hotspots and their sensitivity to topography. This preliminary action identifies which regions require detailed topography measurement, enabling targeted data collection that maximizes hotspot identification accuracy while minimizing information loss
Data Source
AI summary
A method of topography determination, the method including: obtaining a first focus value derived from a computational lithography model modeling patterning of an unpatterned substrate or derived from measurements of a patterned layer on an unpatterned substrate; obtaining a second focus value derived from measurement of a substrate having a topography; and determining a value of the topography from the first and second focus values.