Model-Based Scanner Tuning for Lithography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scanner tuning methods for lithography systems are inefficient and time-consuming, particularly due to the need for trial and error processes to match imaging performance across different scanners, which often results in inconsistent results for two-dimensional patterns despite adequate one-dimensional pattern matching.
Innovation Solution
The development of model-based systems and methods that utilize sensitivity and differential models to simulate and optimize scanner settings, allowing for full-chip level simulations and verification, enabling precise tuning of scanners to match reference performance by accounting for deviations and variances in imaging behavior across multiple scanners.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If trial and error process is used to tune scanner settings, then scanner can be adjusted to achieve acceptable imaging results, but the process becomes time-consuming and expensive
Solution Approach 1:
The patent applies preliminary action by pre-characterizing each scanner's optical proximity effects (OPEs) using measurement tools before actual production tuning. Scanner-specific OPE parameters are measured and stored in a database, creating a knowledge base that guides subsequent tuning operations. This preliminary characterization eliminates the need for extensive trial-and-error adjustments during production, significantly reducing tuning time while maintaining imaging quality.
2Manufacturing precision
If proximity matching is performed on one-dimensional patterns, then critical dimension uniformity is improved, but two-dimensional pattern matching remains inconsistent
Solution Approach 1:
The patent extends the tuning approach from one-dimensional patterns to two-dimensional patterns by incorporating multiple pattern types (1D lines, 2D contacts, mandrels, and real device patterns) into the OPE characterization process. The system measures and models OPEs across different pattern dimensions and uses this comprehensive model to predict and correct imaging behavior for any pattern type, ensuring consistent matching across both 1D and 2D patterns.
Solution Approach 2:
The patent changes the approach from adjusting physical scanner parameters through trial-and-error to modifying computational parameters in a lithography model. By adjusting model parameters (such as optical coefficients and proximity effect parameters) based on measured OPE data, the system achieves accurate pattern prediction and matching without physical scanner reconfiguration, improving both 1D and 2D pattern consistency.
3Reliability
If multiple patterns are measured for tuning, then comprehensive matching is achieved, but wafer metrology time is excessively consumed
Solution Approach 1:
The patent uses computational modeling to create virtual copies of pattern imaging behavior instead of physically measuring every pattern on every wafer. A lithography model is constructed that replicates the scanner's imaging characteristics based on initial measurements. This model can then predict imaging results for any pattern without requiring actual wafer measurement, dramatically reducing metrology time while maintaining matching accuracy.
Solution Approach 2:
The system performs preliminary OPE characterization using a representative set of patterns to build a comprehensive scanner model. Once this model is established, it can predict imaging behavior for any additional patterns without requiring further wafer measurements. This preliminary modeling approach achieves comprehensive matching accuracy while minimizing actual wafer metrology time to only what's needed for initial characterization.
Data Source
AI summary
Systems and methods for tuning photolithographic processes are described. A model of a target scanner is maintained defining sensitivity of the target scanner with reference to a set of tunable parameters. A differential model represents deviations of the target scanner from the reference. The target scanner may be tuned based on the settings of the reference scanner and the differential model. Performance of a family of related scanners may be characterized relative to the performance of a reference scanner. Differential models may include information such as parametric offsets and other differences that may be used to simulate the difference in imaging behavior.