Constant-Width SRAFs for Sub-Resolution Lithography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lithographic projection systems face challenges in reproducing patterns with dimensions smaller than the classical resolution limit, leading to difficulties in achieving precise electrical functionality and performance, especially with the increasing complexity of semiconductor manufacturing.
Innovation Solution
The method involves generating mask patterns with sub-resolution assist features (SRAFs) of constant widths, which are optimized during a mask optimization process using a lithography model and cost function, incorporating optical proximity correction and source-mask co-optimization to improve imaging characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of moving object
If lithographic projection systems attempt to reproduce patterns with dimensions smaller than the classical resolution limit, then the ability to print sub-resolution features is improved, but manufacturing precision deteriorates due to diffraction effects and inability to achieve faithful pattern transfer
Solution Approach 1:
The patent introduces sub-resolution assist features (SRAFs) as intermediary elements that do not print directly but modify the optical field to enable faithful reproduction of smaller main features. These SRAFs act as mediators between the illumination source and the main pattern features, using diffraction effects constructively to improve image contrast and resolution for features below the classical resolution limit.
Solution Approach 2:
The patent optimizes multiple parameters including SRAF width, SRAF spacing, SRAF position, illumination numerical aperture (NA), and wavelength to push the resolution limit. By systematically varying these parameters and using optimization algorithms, the system achieves improved pattern fidelity for sub-resolution features while maintaining manufacturing precision.
2Adaptability or versatility
If the complexity of semiconductor manufacturing patterns increases, then device functionality is improved, but the difficulty of achieving precise pattern transfer worsens due to increased optical interference and proximity effects
Solution Approach 1:
The patent applies local quality by placing SRAFs selectively at specific locations around main features where optical interference most affects pattern fidelity. The SRAF presence, width, and position are locally optimized for each feature based on its specific geometric context, allowing complex patterns to be printed with high precision despite varying local optical conditions.
Solution Approach 2:
The patent employs iterative optimization algorithms that use feedback from lithography simulations to adjust SRAF parameters. The process simulates the lithographic printing, evaluates pattern fidelity, and refines SRAF dimensions and positions to minimize edge placement errors and maximize pattern accuracy for complex device structures.
3Ease of manufacture
If constant-width SRAFs are used instead of variable-width SRAFs, then ease of manufacture is improved, but manufacturing precision may deteriorate due to reduced optimization flexibility
Solution Approach 1:
The patent systematically studies the impact of constraining SRAF width to constant values versus allowing variable widths. By optimizing other parameters (SRAF position, spacing, illumination conditions) under the constant-width constraint, the patent determines that sufficient pattern fidelity can be achieved while dramatically simplifying mask fabrication processes and reducing manufacturing costs.
Data Source
AI summary
Methods, software, and systems are disclosed for determining mask patterns. The determination can include obtaining a mask pattern including sub-resolution assist features (SRAFs) each having constant widths. The widths are set as continuous variables and so can be optimized along with other variables during a mask optimization process of the mask pattern. Based on their population and/or statistics, the optimized continuous widths are then discretized to a limited number of global width levels. Further mask optimization be performed with the SRAFs having discretized optimized global width levels, where the width assigned to an individual SRAF may be adjusted to a different level of the global width levels.