Computational Metrology for Lithographic Overlay Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current device manufacturing processes face challenges in accurately determining and correcting for alignment and overlay errors in lithographic apparatuses, leading to inefficiencies and defects in semiconductor production.
Innovation Solution
A method involving a hardware computer system that determines distributions of alignment data and error contributions, combining measured alignment data with processing parameters to estimate overlay and alignment errors, allowing for precise correction and improvement of the patterning process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional alignment and overlay measurement methods are used in lithographic apparatus, then the measurement process is simplified, but the measurement precision and accuracy of alignment errors are insufficient
Solution Approach 1:
The patent performs preliminary computational analysis by determining distributions of multiple parameters (focus, exposure dose, overlay errors, alignment errors) before final measurement results are obtained. This allows the system to predict and correct for various error contributions in advance, improving measurement precision without requiring more complex hardware during the actual measurement process.
Solution Approach 2:
The patent introduces computational models and simulation data as intermediaries between the physical measurement process and the final alignment results. By using computational metrology to model and separate different error contributions (lens aberrations, focus errors, overlay errors, alignment errors), the system achieves higher precision without directly complicating the physical measurement apparatus.
2Manufacturing precision
If comprehensive error analysis with multiple parameter distributions is performed, then the manufacturing precision is improved, but the loss of time in processing increases
Solution Approach 1:
The patent performs comprehensive error distribution analysis as a preliminary step before actual patterning operations. By determining distributions of focus, exposure dose, overlay errors, and alignment errors in advance, the system establishes correction models that can be applied rapidly during production, achieving high manufacturing precision without repeating the full computational analysis for each part.
Solution Approach 2:
The patent creates computational models and simulated error distributions that replicate the complex relationships between various process parameters. These models serve as copies of the physical system's error behavior, allowing the system to predict and correct errors without performing time-consuming physical measurements and analyses for each individual case.
3Reliability
If advanced computational metrology methods are implemented, then the reliability of the patterning process is improved, but the device complexity increases
Solution Approach 1:
The patent segments the complex error analysis into distinct components by determining separate distributions for different parameters (focus errors, exposure dose variations, overlay errors, alignment errors). This segmentation allows each error source to be analyzed and corrected independently, improving reliability while managing computational complexity through modular processing of individual error types.
Solution Approach 2:
The patent employs computational models as intermediaries that mediate between the complex physical patterning process and the control system. These models translate complex multi-parameter error distributions into actionable correction data, enhancing process reliability without requiring the physical apparatus itself to become more complex.
Data Source
AI summary
A method, involving determining a first distribution of a first parameter associated with an error or residual in performing a device manufacturing process; determining a second distribution of a second parameter associated with an error or residual in performing the device manufacturing process; and determining a distribution of a parameter of interest associated with the device manufacturing process using a function operating on the first and second distributions. The function may include a correlation.