Focus Prediction Map for Multi-Layer Lithography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In lithographic apparatuses, accurately compensating for focus deviations across substrates with multiple layers is challenging, as traditional methods struggle to account for distortions and variations in wafer thickness.
Innovation Solution
The method involves obtaining key performance indicator data from alignment marks using an alignment sensor, determining a correlation between this data and focus offset data, and generating a focus prediction map to compensate for focus deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional focus compensation methods are used, then the process is simple, but the manufacturing precision deteriorates due to inability to account for multi-layer distortions and thickness variations
Solution Approach 1:
The substrate is divided into multiple measurement points across different layers, with each point providing localized height and alignment data. This segmentation allows the system to capture layer-specific variations rather than treating the substrate as a uniform entity, thereby improving focus compensation precision for multi-layer structures.
Solution Approach 2:
The system performs preliminary measurements of alignment marks and height maps on all substrate layers before the actual lithographic exposure. This preliminary action generates a focus prediction map that is used to pre-compensate for distortions, ensuring that when exposure occurs, the focus is already optimized for each region's specific topography.
2Measurement precision
If alignment marks are measured for coordinate system derivation, then the position accuracy is improved, but substrate distortion from clamping and processing deteriorates the measurement accuracy
Solution Approach 1:
The system measures and maps substrate distortions at multiple layers before the final lithographic exposure. By creating a distortion map in advance, the system can compensate for these harmful effects during the exposure process, ensuring that alignment measurements remain accurate despite the presence of substrate distortions from clamping and processing.
3Manufacturing precision
If focus prediction maps are generated for each layer, then the overlay accuracy is improved, but the processing time and complexity increase
Solution Approach 1:
The system generates focus prediction maps for all layers in advance, before the actual lithographic exposure. This preliminary generation of correction maps allows the exposure process to proceed with pre-calculated focus adjustments, improving overlay accuracy without adding time during the critical exposure step itself.
Solution Approach 2:
The system combines height map data and alignment mark measurements from multiple layers into a single integrated focus prediction map. This merging of data sources creates a comprehensive correction model that accounts for inter-layer variations, improving overlay accuracy while streamlining the processing through unified map generation.
Data Source
AI summary
A method of compensating for focus deviations on a substrate having a plurality of layers present thereon, the method includes generating a focus prediction map for the substrate. In one approach, the focus prediction map is generated by obtaining key performance indicator data on the substrate using an alignment sensor, determining a correlation between the KPI data and focus offset data for positions on the substrate, and using the correlation and the KPI data, generating a focus prediction map for the substrate. In another approach, the prediction map is generated by obtaining a first layer height map for a first layer, measuring, with a level sensor, a second layer height map for a second layer overlying the first layer, and subtracting the first height map from the second height map to obtain a delta height map for the substrate.