Adaptive Semiconductor Wafer Metrology Sampling Plan
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face inefficiencies in sampling plans for wafer metrology, which affect the effectiveness of process control and yield improvement, as they do not adequately adapt to changes in equipment states or tool signatures.
Innovation Solution
A method is introduced that defines a sampling plan with position information for metrology sites on process wafers, where measurement values from high-resolution metrology tools are used to update a sampling model, allowing for adaptive adjustments to the sampling plan based on deviations from the current model, thereby focusing on regions of high interest and minimizing those of low interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a traditional sampling plan with fixed metrology sites is used, then the measurement process is simple and fast, but the measurement precision and reliability are insufficient to detect process deviations
Solution Approach 1:
The sampling plan transitions from a static, fixed configuration to a dynamic, adaptive system. The metrology site positions and the number of sites are continuously adjusted based on real-time process data and deviation assessments, allowing the measurement system to adapt its complexity to the actual process needs while maintaining high measurement precision.
Solution Approach 2:
The measurement system performs self-optimization by automatically assessing process deviations and autonomously updating the sampling plan without external intervention. The system uses its own measurement data to identify regions of interest and dynamically reconfigure metrology sites, enabling self-improving measurement precision.
2Reliability
If metrology sites are increased to improve process control, then the reliability of process information improves, but the measurement time and productivity decrease
Solution Approach 1:
Instead of uniformly increasing metrology sites across the entire wafer, the system concentrates measurement resources in locally identified regions of interest where process deviations occur. The sampling plan dynamically adjusts site positions and densities based on assessed deviation patterns, ensuring high reliability in critical areas while maintaining productivity through reduced measurements in stable areas.
Solution Approach 2:
The system applies partial action by measuring only the necessary portions of the wafer at any given time. Rather than performing exhaustive measurements across all sites, the adaptive sampling plan selectively targets regions showing deviations, achieving sufficient reliability for process control while minimizing measurement time and maintaining productivity.
3Adaptability or versatility
If the sampling plan is updated frequently to adapt to equipment state changes, then the adaptability improves, but the complexity of plan management and processing time increase
Solution Approach 1:
The system implements continuous feedback loops where measurement results are immediately assessed against process specifications, and sampling plan updates are triggered only when deviations exceed predefined thresholds. This feedback-driven approach enables the system to adapt to equipment state changes in real-time while avoiding unnecessary plan updates, thus maintaining high adaptability without excessive processing time or management complexity.
Solution Approach 2:
The sampling plan parameters (site positions, number of sites, measurement frequencies) are dynamically changed based on assessed process deviations. The system monitors equipment state changes and adjusts sampling parameters accordingly, enabling flexible adaptation while minimizing the time and effort required for plan management through automated, data-driven parameter optimization.
4Loss of information
If comprehensive measurements are performed across all wafer regions, then the completeness of process characterization is high, but the measurement effort and costs increase significantly
Solution Approach 1:
The wafer measurement process is segmented into distinct regions based on assessed process deviations. Instead of treating the entire wafer uniformly, the system divides it into regions of interest (showing deviations) and regions of stability, applying measurement resources selectively to each segment. This ensures complete characterization of problematic areas while avoiding unnecessary measurements in stable areas, reducing overall measurement effort and energy consumption.
Solution Approach 2:
The system extracts and focuses measurement efforts on the essential information needed for effective process control. By identifying and isolating regions with process deviations, the system extracts the critical data required for characterization while eliminating redundant measurements in stable regions, thereby maintaining complete process understanding with reduced measurement effort and energy use.
Data Source
AI summary
A method of manufacturing semiconductor devices includes defining a sampling plan that contains position information about metrology sites on process wafers. A first property of the process wafers is measured to obtain measurement values at measurement points, wherein a quantity of the measurement points per process wafer is at least tenfold a quantity of the metrology sites. A sampling model that includes at least a wafer model is updated on the basis of the measurement values. The sampling plan is updated on the basis of an assessment of deviations of the measurement values from a current sampling model.