Metrology Target Optimization for Semiconductor Process Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lithographic processes face challenges in accurately monitoring and controlling the manufacturing of semiconductor devices due to variations in metrology target designs and recipes, leading to inefficiencies in process optimization and quality control.
Innovation Solution
A method involving simulation and data-driven approaches to evaluate and identify optimal metrology targets and recipes, using measurement data to tune parameters and adjust sampling schemes, thereby improving process control and precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple different metrology target designs and recipes are used to monitor manufacturing processes, then process monitoring coverage is improved, but measurement precision deteriorates due to variability across different targets and recipes
Solution Approach 1:
The patent applies parameter changes by systematically varying metrology target design parameters (such as target geometry, material composition, and structural features) and recipe parameters (such as measurement conditions and processing parameters) to identify optimal combinations. This allows the system to adapt to different manufacturing scenarios while maintaining consistent measurement precision through data-driven optimization of these parameters
Solution Approach 2:
The patent implements feedback mechanisms by using measurement data from multiple metrology targets and recipes to continuously refine and optimize the measurement process. The collected data is analyzed to identify variations and trends, which then feed back into adjusting target designs and measurement recipes to improve both coverage and precision iteratively
2Measurement precision
If simulation is used to evaluate multiple metrology targets and recipes, then measurement precision is improved, but loss of time increases due to extensive simulation and data analysis
Solution Approach 1:
The patent applies preliminary action by performing simulations and evaluations of multiple metrology target designs and recipes before actual manufacturing measurements. This advance evaluation identifies the most promising target-recipe combinations, so that when actual manufacturing occurs, the optimized parameters can be applied directly, reducing the need for extensive real-time simulation and analysis
Solution Approach 2:
The patent applies partial action by selectively evaluating only the most critical and influential parameters through simulation, rather than exhaustively analyzing all possible parameters. This focused approach identifies the key factors affecting measurement precision while significantly reducing the computational time and resources required
Data Source
AI summary
A method including evaluating, with respect to a parameter representing remaining uncertainty of a mathematical model fitting measured data, one or more mathematical models for fitting measured data and one or more measurement sampling schemes for measuring data, against measurement data across a substrate, and identifying one or more mathematical models and/or one or more measurement sampling schemes, for which the parameter crosses a threshold.