Lithography Metrology Modeling for Cross-Tool Overlay Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lithographic processes face challenges in accurately modeling and correcting distortions in substrate grids, particularly when different processing tools are used, leading to misalignments and overlay errors in semiconductor manufacturing, which are not adequately addressed by current advanced alignment models.
Innovation Solution
A method for determining process corrections by obtaining metrology data, modeling it using a first model related to a specific apparatus, and distributing penalties across processes to ensure performance parameters meet respective specifications, incorporating a computer program to perform these operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If different apparatuses with varying correction actuation capabilities are used in lithographic processes, then process flexibility and adaptability are improved, but overlay precision deteriorates due to inability to uniformly meet overlay specifications across all processes
Solution Approach 1:
The patent changes the parameters of the alignment model to match the correction actuation capability of different apparatuses. By adjusting model parameters rather than requiring uniform correction capability, the system can adapt to various apparatus capabilities while still meeting overlay specifications through optimized penalty distribution.
Solution Approach 2:
The patent segments the overlay penalty into process-specific portions that can be independently optimized. Each apparatus process receives a customized penalty allocation based on its correction capability, allowing the system to handle diverse apparatus capabilities while maintaining overall overlay precision.
2Ease of manufacture
If overlay penalties are not distributed across processes, then individual process optimization is simplified, but overall overlay specifications cannot be met when apparatus correction capabilities differ
Solution Approach 1:
The system introduces process-specific penalty parameters that can be independently adjusted. By changing the penalty parameters for each process based on apparatus capability, the system achieves both simplified individual optimization and overall specification compliance through coordinated parameter adjustment.
Solution Approach 2:
The penalty distribution is made dynamic and adaptable to each apparatus's correction capability. Rather than a fixed penalty structure, the system dynamically allocates penalties based on actual apparatus performance characteristics, enabling both ease of individual optimization and overall compliance.
3Device complexity
If a single alignment model is used for all apparatuses, then model complexity is reduced, but accuracy deteriorates when apparatuses have different correction actuation capabilities
Solution Approach 1:
The patent applies local quality by customizing alignment model parameters for each apparatus based on its specific correction actuation capability. Each apparatus receives a locally optimized model configuration rather than a uniform model, improving alignment accuracy while maintaining manageable overall system complexity through modular parameter adjustment.
Data Source
AI summary
Disclosed is a method for determining a process correction for at least a first process of a lithographic process, comprising at least the first process performed on at least a first substrate using at least a first apparatus and a second process performed on at least said first substrate using at least a second apparatus, where a correction actuation capability of the first apparatus differs from the second apparatus, comprising: obtaining metrology data relating to said first substrate; modeling said metrology data using a first model, the model being related to said first apparatus; and controlling said first process based on the modeled metrology data; the modeling step and/or an additional processing step comprises distributing a penalty in a performance parameter across said first process and said second process such that the distributed penalties in the performance parameter are within their respective specifications of the performance parameter.