Pattern-Dependent OPC Model Selection for Semiconductor Lithography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical proximity correction (OPC) systems face challenges in accurately correcting pattern distortions due to the optical proximity effect, especially in high-density semiconductor devices, as they often rely on generic OPC models that fail to account for specific pattern shapes and dispositions, leading to significant fitting errors and inefficiencies.
Innovation Solution
An OPC system that generates an integrated circuit (IC) layout and selects specific OPC models based on target characteristics, applying different models to various portions of the layout to generate a mask layout that accurately reflects lithography processing effects, thereby improving pattern correction and reducing fitting errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a generic OPC model is applied to the entire layout, then the processing is simple and fast, but the accuracy of pattern correction deteriorates due to fitting errors from not accounting for specific pattern shapes and dispositions
Solution Approach 1:
The patent divides the IC layout into multiple fragments and classifies them into different types (e.g., line-end patterns, line-and-space patterns, block patterns) based on their geometric characteristics. Each fragment type is associated with a specific OPC model from a database, allowing tailored correction for each pattern type while maintaining overall system organization through systematic classification.
Solution Approach 2:
The patent applies different OPC models to different portions of the layout based on their specific pattern characteristics. Instead of using a single generic model for the entire layout, the system selects and applies the most appropriate model for each fragment type, such as using line-end treatment models for patterns with serifs and scattering bar models for patterns requiring density compensation, thereby optimizing correction accuracy for each local region.
2Manufacturing precision
If multiple specific OPC models are selected and applied to different layout portions, then the pattern correction accuracy improves, but the processing time and computational complexity increase
Solution Approach 1:
The patent pre-classifies layout fragments into different types and pre-associates them with appropriate OPC models from a database before the actual correction process. This preliminary organization allows the system to quickly retrieve and apply the correct model without performing complex real-time calculations, significantly reducing processing time while maintaining high accuracy.
Solution Approach 2:
The patent implements a dynamic model selection mechanism where the appropriate OPC model is chosen based on the specific characteristics of each layout fragment. The system can adaptively select from multiple predefined models (e.g., line-end treatment, scattering bar insertion, basic OPC) depending on the fragment type, allowing flexible and optimized processing that adjusts to different pattern characteristics without requiring exhaustive computation for all possible cases.
3Productivity
If rule-based OPC processes are used, then the processing speed is fast, but the ability to handle diverse pattern configurations deteriorates due to difficulty in establishing valid rules for all possible mask transfers
Solution Approach 1:
The patent uses a database of pre-characterized OPC models that have been developed and validated for specific pattern types. Instead of creating new rules for each possible pattern configuration, the system copies and applies appropriate pre-existing models to different layout fragments. This approach maintains the speed advantage of rule-based processing while achieving versatility by selecting from a comprehensive database of proven models that cover various pattern configurations.
Data Source
AI summary
An optical proximity correction (OPC) system and methods thereof are provided. The example OPC system may include an integrated circuit (IC) layout generation unit generating an IC layout, a database unit storing a first plurality of OPC models, each of the first plurality of OPC models associated with one of a plurality of target specific characteristics and a mask layout generation unit including a model selector selecting a second plurality of OPC models based on a comparison between the target specific characteristics associated with the plurality of OPC models and the generated IC layout, the mask layout generation unit generating a mask layout based on the IC layout and the selected second plurality of OPC models. A first example method may include storing a first plurality OPC models, each of the first plurality of OPC models associated with one of a plurality of target specific characteristics, generating an IC layout, selecting a second plurality of OPC models based on a comparison between the target specific characteristics associated with the first plurality of OPC models and the generated IC layout and generating a mask layout based on the generated IC layout and the selected second plurality of OPC models. A second example method may include applying a first OPC model to a first portion of a generated integrated circuit (IC) layout, applying a second OPC model to a second portion of the generated IC layout and generating a mask layout based on the generated IC layout after the application of the first and second OPC models.