Scanner Overlay Correction Using Regularized Regression Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor circuit line widths become finer, overlay errors between patterns of different layers occur due to the complexity of extreme ultra-violet (EUV) exposure processes, leading to suboptimal semiconductor device performance.
Innovation Solution
An overlay correcting method using regularized regression to calculate correction parameters within the scanner's correction limit, ensuring accurate overlay correction by adjusting hyper-parameters in ridge regression, LASSO regression, or elastic net regression to prevent over-fitting and maintain parameters within the scanner's capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If regular regression is used to calculate overlay correction parameters, then overlay correction capability is maximized, but correction parameters may exceed the scanner's correction limit
Solution Approach 1:
The patent transforms the regression approach from standard regression to regularized regression (ridge, LASSO, elastic net), changing the mathematical parameters and constraints of the calculation method. This parameter change ensures that correction parameters remain within the scanner's correction limit while maintaining overlay correction effectiveness
Solution Approach 2:
The patent performs preliminary adjustment of correction parameters through regularized regression before they are applied to the scanner. By pre-calculating parameters that satisfy both overlay correction requirements and scanner limitations, the system avoids the problem of exceeding correction limits during actual operation
2Manufacturing precision
If standard regression is applied without constraints, then overlay correction is optimized, but over-fitting occurs reducing reliability
Solution Approach 1:
The patent introduces regularization parameters (alpha, lambda) that control the balance between fitting the training data and maintaining model generalization. By adjusting these parameters, the system prevents over-fitting while maintaining effective overlay correction capability
Solution Approach 2:
The patent incorporates feedback mechanisms through cross-validation and performance evaluation metrics to select optimal regularization parameters. This feedback loop ensures that the correction model generalizes well to unseen data while maintaining accuracy on training data
Data Source
AI summary
An overlay correcting method capable of optimizing correction of an overlay within a scanner correction limit of a scanner of a scanner system, and a photolithography method, a semiconductor device manufacturing method and the scanner system which are based on the overlay correcting method are provided. The overlay correcting method includes collecting overlay data by measuring an overlay of a pattern; calculating correction parameters of the overlay by performing regularized regression using the overlay data, the regularized regression being based on a correction limit of the scanner such that the correction parameters fall within the correction limit of the scanner; and providing the correction parameters to the scanner.