Photomask Layout Correction for EUV Pattern Fidelity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mask manufacturing technologies face challenges in achieving enhanced pattern fidelity and consistency for advanced integrated circuit technology nodes, particularly in EUVL and optical lithography, due to issues like critical dimension uniformity, corner rounding, and sub-resolution assist features, which conventional tools struggle to address effectively.
Innovation Solution
A system and method utilizing Litho Aware Mask Process correction Application (LAMA) techniques, involving spatial domain analysis and SEM image contour extraction, to enhance photomask manufacturing by identifying and correcting defects through a multi-step data flow that harmonizes mask and wafer processes, applying corrective photomask structures to improve pattern fidelity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional mask manufacturing tools and methods are used, then manufacturing simplicity is maintained, but pattern fidelity and manufacturing precision deteriorate at advanced technology nodes
Solution Approach 1:
The mask manufacturing process is segmented into multiple distinct stages: initial mask creation, SEM imaging, contour extraction, defect analysis, corrective structure generation, and verification. Each stage produces specific data that feeds into the next, allowing complex corrections to be applied systematically without overwhelming the manufacturing system
Solution Approach 2:
Corrective photomask structures are generated and applied before final mask manufacturing. The system performs preliminary defect analysis on SEM images, generates correction patterns, and incorporates them into the mask design before production, preventing defect propagation to the final product
Solution Approach 3:
The system creates digital copies of mask contours from SEM images and generates virtual corrective structures in the digital domain. These digital corrections are then applied to the mask layout before physical manufacturing, allowing verification and optimization without affecting actual mask production
2Reliability
If mask process corrections are applied to enhance pattern fidelity, then lithographic performance improves, but process time and complexity increase
Solution Approach 1:
The system performs automated defect detection, analysis, and correction generation without requiring extensive manual intervention. Algorithms automatically extract contours from SEM images, identify defect patterns, generate corrective structures, and apply them to mask layouts, reducing both time and human resource requirements
Solution Approach 2:
The system uses SEM imaging to obtain feedback on actual mask structures, compares them against design specifications, and generates corrections based on detected deviations. This closed-loop feedback mechanism ensures corrections are targeted and effective, reducing the number of iterative cycles needed
3Manufacturing precision
If conventional single-beam writing tools are used, then device simplicity is maintained, but manufacturing precision deteriorates for EUVL requirements
Solution Approach 1:
The system changes the operational parameters of conventional single-beam writing tools by implementing multi-pass writing strategies and dynamic parameter adjustment during mask manufacturing. This allows existing tools to achieve EUVL-level precision without requiring complete tool replacement
Solution Approach 2:
The mask manufacturing process incorporates dynamic parameter adjustment where writing conditions, beam parameters, and correction application timing are optimized in real-time based on detected defect patterns and structure complexity, enabling conventional tools to adapt to advanced precision requirements
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Methods and systems for building a photomask from obtained pattern information relating to a photomask that exhibits defects on wafer. Spatial domain analysis is conducted on the pattern information so that corrective photomask structures can be generated and applied to a photomask layout. A photomask is built using the corrective photomask structures. The photomask is verified for effectiveness.