ML-Based Optical Proximity Correction for Faster Mask Layout Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lithographic projection technologies face challenges in accurately reproducing patterns with dimensions smaller than the classical resolution limit, requiring complex and time-consuming optical proximity correction (OPC) processes that are computationally intensive and struggle with runtime and accuracy.
Innovation Solution
A machine learning model is trained to predict post-OPC images by using pre-OPC images and assist features, reducing the difference with reference images through iterative weight adjustments, enabling faster and more accurate OPC simulations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional optical proximity correction (OPC) processes are used to accurately reproduce patterns smaller than the classical resolution limit, then manufacturing precision is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent replaces traditional iterative OPC computational algorithms with a machine learning model that has been trained to predict post-OPC layouts directly from pre-OPC layouts. This substitution transforms the complex iterative computational process into a more efficient predictive model, maintaining high pattern reproduction accuracy while significantly reducing computational complexity and processing time.
2Manufacturing precision
If traditional iterative OPC processes are used to achieve accurate pattern correction, then manufacturing precision is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by training the machine learning model in advance using extensive OPC training data and reference layouts. Once trained, the model can rapidly predict post-OPC layouts without requiring time-consuming iterative corrections during actual production, thus maintaining high accuracy while dramatically reducing processing time.
3Manufacturing precision
If complex iterative OPC algorithms are used to handle assist features and proximity effects, then manufacturing precision is improved, but productivity decreases
Solution Approach 1:
The patent substitutes traditional iterative OPC algorithms with a trained machine learning model that can process multiple layouts simultaneously. This substitution maintains high correction accuracy for assist features and proximity effects while significantly improving processing throughput and productivity.
Data Source
AI summary
Training methods and a mask correction method. One of the methods is for training a machine learning model configured to predict a post optical proximity correction (OPC) image for a mask. The method involves obtaining (i) a pre-OPC image associated with a design layout to be printed on a substrate, (ii) an image of one or more assist features for the mask associated with the design layout, and (iii) a reference post-OPC image of the design layout; and training the machine learning model using the pre-OPC image and the image of the one or more assist features as input such that a difference between the reference image and a predicted post-OPC image of the machine learning model is reduced.