Metrology Offset Prediction Using Spatial Wafer Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the metrology-to-device offset in semiconductor manufacturing are limited, often relying on simplistic assumptions that do not fully capture spatial variations, leading to suboptimal process control and reduced yield.
Innovation Solution
A method using a trained machine learning model to predict spatially varying process offsets by correlating different types of metrology data, such as overlay and electrical characteristic data, to provide a spatially varying offset map that can be used to improve process control and yield optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional optical measurement methods are used to determine metrology-to-device offset, then measurement simplicity is maintained, but measurement precision deteriorates due to spatial variation assumptions
Solution Approach 1:
The patent changes the measurement parameters by using multiple measurement types (optical overlay measurements at different wavelengths, electron beam measurements) instead of a single measurement type. This allows capturing spatial variations in the metrology-to-device offset across the wafer surface, improving measurement precision without requiring a single complex measurement system
Solution Approach 2:
The patent employs a multi-functional measurement approach where different measurement tools (optical metrology tools, electron beam tools) are used to measure the same physical quantity (metrology-to-device offset) but with different characteristics. This universal measurement strategy improves precision by combining multiple measurement perspectives while managing complexity through systematic data processing
2Manufacturing precision
If spatially varying offset correction is implemented, then manufacturing precision improves, but process control complexity increases
Solution Approach 1:
The patent applies local quality by determining spatially varying metrology-to-device offset values across different locations on the wafer surface. Instead of using a single global offset value, the system calculates location-specific offset values that account for local variations in the lithographic process, thereby improving manufacturing precision. The complexity is managed by using systematic measurement and modeling approaches
Solution Approach 2:
The patent segments the wafer surface into multiple measurement locations and determines offset values for each location independently. This segmentation allows the process control system to handle complexity in a manageable way by processing location-specific data separately, then combining the results to achieve overall improved manufacturing precision
3Measurement precision
If multiple measurement types are correlated to predict offset, then measurement precision improves, but loss of time increases due to additional measurements
Solution Approach 1:
The patent performs preliminary measurements using optical metrology tools at multiple locations before final offset determination. These preliminary measurements provide initial offset estimates that guide subsequent more precise measurements, reducing the total time required while maintaining high measurement precision through staged measurement approaches
Solution Approach 2:
The patent uses optical measurements as a proxy or copy of the more time-consuming electron beam measurements. By correlating the faster optical measurement data with electron beam data, the system predicts offset values with high precision without requiring time-consuming direct electron beam measurements at all locations
Data Source
AI summary
A method for determining a spatially varying process offset for a lithographic process, the spatially varying process offset (MTD) varying over a substrate subject to the lithographic process to form one or more structures thereon. The method includes obtaining a trained model (MOD), having been trained to predict first metrology data based on second metrology data, wherein the first metrology data (OV) is spatially varying metrology data which relates to a first type of measurement of the one or more structures being a measure of yield and the second metrology data (PB) is spatially varying metrology data which relates to a second type of measurement of the one or more structures and correlates with the first metrology data; and using the model to obtain the spatially varying process offset (MTD).