Lithographic Process Co-Optimization for Pattern Fidelity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lithographic processes face challenges in accurately reproducing small feature sizes and high feature densities due to proximity effects, which are not adequately addressed by existing resolution enhancement techniques (RET) such as optical proximity correction (OPC), especially in deep sub-wavelength regimes where feature dimensions are sensitive to local environments and neighboring features.
Innovation Solution
A computer-implemented method for simultaneously optimizing design variables of the pattern transfer process and design rules in lithographic projection apparatuses, using a cost function to derive optimal values for parameters like illumination source shape, mask bias, and projection optics, to achieve a favorable process window and minimize errors in critical dimension uniformity across varying process conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional optical proximity correction (OPC) is used to enhance resolution, then larger feature sizes can be printed, but small feature sizes and high feature densities cannot be accurately reproduced due to proximity effects
Solution Approach 1:
The patent applies preliminary action by pre-distorting the mask pattern to compensate for anticipated proximity effects during lithography. The system calculates correction values based on the local environment of each feature and applies these corrections before the actual printing process, allowing small features to be accurately reproduced despite sensitivity to process variations
Solution Approach 2:
The patent implements local quality by applying different correction strategies to different features based on their local environment. Each feature's mask pattern is individually adjusted according to its surrounding features, enabling accurate reproduction of small features while maintaining overall pattern fidelity across varying process conditions
2Productivity
If design rules are optimized for small feature sizes, then feature density increases, but proximity effects from neighboring features cause printing errors
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting mask pattern parameters based on local feature density and proximity relationships. The system modifies line widths, spacing, and other geometric parameters to compensate for proximity effects, enabling high feature density while maintaining critical dimension uniformity across the wafer
Solution Approach 2:
The patent implements feedback through an iterative optimization process that evaluates the impact of each feature on its neighbors and adjusts the mask pattern accordingly. The system calculates proximity effects from neighboring features and uses this information to refine the mask design, achieving both high density and uniformity through repeated refinement cycles
3Manufacturing precision
If resolution enhancement techniques are applied to maintain feature fidelity, then pattern accuracy improves, but the process window narrows due to sensitivity to process variations
Solution Approach 1:
The patent merges multiple optimization objectives into a unified cost function that simultaneously considers pattern fidelity, process window, and proximity effects. By combining these previously separate optimization goals into a single framework, the system achieves high pattern accuracy while maintaining a robust process window without requiring separate correction passes
Data Source
AI summary
A computer-implemented method for obtaining values of one or more design variables of one or more design rules for a pattern transfer process comprising a lithographic projection apparatus, the method comprising: simultaneously optimizing one or more design variables of the pattern transfer process and the one or more design variables of the one or more design rules. The optimizing comprises evaluating a cost function that measures a metric characteristic of the pattern transfer process, the cost function being a function of one or more design variables of the pattern transfer process and one or more design variables of the one or more design rules.