EUV Mask Defect Alignment via Non-Linear Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current EUV lithography technologies face challenges in producing defect-free mask blanks and lack cost-effective tools to repair buried defects in EUV masks, which impact pattern quality and increase wastage of expensive EUV blanks.
Innovation Solution
A system and method using a cost function based on defect exposure area and pattern complexity, optimized through non-linear procedures, to adjust the rotation and translation of mask blanks, effectively hiding defects under the absorber layer, thereby minimizing their impact on pattern printability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mask blanks are inspected and defects are identified, then defect visibility is improved, but the ability to repair buried defects remains insufficient
Solution Approach 1:
The system performs preliminary inspection and identification of defects in the mask blank before the patterning process. By detecting defects early and calculating optimal alignment parameters, the system prepares compensation strategies in advance, allowing the design pattern to be positioned to minimize defect impact without requiring actual repair of the buried defects.
2Ease of manufacture
If the design pattern is fixed on the mask blank, then manufacturing simplicity is maintained, but defect impact on pattern printability increases
Solution Approach 1:
The system introduces dynamic alignment by calculating optimal rotation and translation parameters that position the design pattern relative to detected defects. This allows the mask to be dynamically adjusted during the alignment process, enabling the pattern to avoid defect regions while maintaining a fixed design on the physical mask blank.
Solution Approach 2:
The system changes the alignment parameters (rotation angle and translation distance) of the design pattern relative to the mask blank based on defect locations. By optimizing these parameters, the system minimizes the impact of defects on pattern printability while keeping the design pattern itself unchanged.
3Loss of substance
If mask blanks with defects are used, then material wastage is reduced, but pattern quality deteriorates
Solution Approach 1:
The system converts the harmful effect of defects into a beneficial outcome by calculating optimal alignment parameters that position the design pattern to minimize defect impact. Defects that would normally render a mask blank unusable are instead managed through intelligent alignment, allowing the mask to be reused and reducing material wastage while maintaining pattern quality.
4Manufacturing precision
If alignment parameters are optimized to hide defects, then pattern printability is improved, but processing time increases
Solution Approach 1:
The system creates a digital model (defect map) of the mask blank defects and uses this copy to calculate optimal alignment parameters through computational optimization. This virtual modeling approach allows rapid evaluation of different alignment scenarios without physical trial-and-error, reducing the time required to determine the optimal pattern positioning.
Data Source
AI summary
A system for reducing printable defects on a pattern mask is disclosed. The system includes a controller configured to be communicatively coupled to a characterization sub-system, the controller including one or more processors configured to execute program instructions causing the one or more processors to: direct the characterization sub-system to perform inspection of a mask blank; generate a cost function based on a first characteristic and a second characteristic, the first characteristic comprising areas of defect regions exposed by mask patterns, the second characteristic comprising pattern complexity of a design pattern; determine one or more values indicative of a minimum of the cost function via a non-linear optimization procedure; and generate one or more control signals to adjust rotation and translation of the mask blank relative to the design pattern based on the determined one or more values indicative of the minimum of the cost function.