Pixel-Based Mask Optimization for Lithography Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As semiconductor device integration increases, the proximity of mask polygons leads to interference and diffraction issues in optical lithography, resulting in distorted circuit patterns, which existing resolution enhancement techniques like optical proximity correction and inverse lithography struggle to address efficiently due to the need for multiple simulations.
Innovation Solution
A method using pixel-based learning to optimize a target mask for a partial coherent system, where a trainer mask is optimized using resolution enhancement techniques, and a mask optimization estimation model is generated by calculating feature vectors based on partial signals from spatial filters, allowing for efficient mask optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If resolution enhancement techniques (optical proximity correction, inverse lithography) are used to optimize mask patterns, then manufacturing precision of circuit patterns is improved, but device complexity and computation time increase due to multiple simulations
Solution Approach 1:
The patent performs mask optimization in advance during the mask design phase, creating an optimized mask pattern before lithography. This preliminary optimization accounts for optical effects and proximity issues, so that the actual lithography process can directly use the pre-optimized pattern without requiring complex real-time simulations during manufacturing.
Solution Approach 2:
The patent creates an optimized mask pattern as a copy or representation that incorporates all necessary corrections for optical effects. This optimized mask serves as a surrogate that reproduces the desired circuit pattern on the substrate without needing to perform complex simulations during the actual lithography process.
2Manufacturing precision
If multiple simulations are performed for mask optimization, then manufacturing precision is improved, but productivity decreases due to increased computation time
Solution Approach 1:
The patent performs the computationally intensive mask optimization simulations in advance during the design phase, creating a finalized optimized mask pattern. This allows complex simulations to be completed before production, enabling fast and accurate lithography during actual manufacturing without compromising productivity.
Solution Approach 2:
The optimized mask pattern serves as a self-contained solution that incorporates all necessary corrections. Once created, the mask can be directly used in lithography without requiring additional simulations or complex real-time computations, allowing the system to serve itself efficiently during production.
3Adaptability or versatility
If mask polygons are placed closer together to increase integration, then device functionality is improved, but optical interference and diffraction increase causing pattern distortion
Solution Approach 1:
The patent applies preliminary anti-action by pre-calculating and compensating for optical interference and diffraction effects during mask design. The mask optimization process anticipates the harmful optical effects that will occur when polygons are placed close together and adjusts the mask pattern in advance to counteract these effects, preventing pattern distortion before lithography.
Solution Approach 2:
The patent changes the parameters of the mask pattern through optimization, adjusting polygon positions, sizes, and shapes to account for optical effects. By modifying these parameters in the mask design phase, the system enables closer polygon placement while maintaining pattern fidelity despite increased optical interference.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables quick and accurate mask optimization for partial coherent systems, reducing the complexity of multiple simulations required by traditional methods and improving the fidelity of printed circuit patterns.
Implementation Method 1
Such proximity can cause interference and diffraction of light
Implementation Method 2
Such proximity can cause interference and diffraction of light
Data Source
AI summary
A mask optimization method for optimizing a target mask used for a partial coherent system including a plurality of spatial filters is provided. The mask optimization method includes obtaining a trainer mask that is an optimized sample mask by performing a mask optimization on a sample mask, generating a mask optimization estimation model by performing a pixel-based learning using, as a feature vector of each of pixels of the trainer mask, partial signals of each of the pixels of the trainer mask respectively determined based on the spatial filters and using, as a target value, a degree of overlap between each of the pixels and a mask polygon of the trainer mask, and performing a mask optimization on the target mask using the mask optimization estimation model.