Robust Optimization for Imprint Lithography Drop Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imprint lithography systems face challenges in achieving precise drop placement and volume distribution due to errors in the electro-mechanical dispense system and environmental variations, leading to non-uniform residual layers and potential surfactant buildup during the nano-fabrication process.
Innovation Solution
The implementation of a robust optimization method using a fluid map and drop volume function to determine optimal drop locations, incorporating techniques such as centroidal Voronoi tessellation and power diagrams, which account for variations in drop placement and volume, ensuring uniformity and minimizing variability through Monte Carlo simulations or structured experiments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional dispense systems are used for applying formable liquid, then the process is simple and fast, but drop placement precision and volume uniformity deteriorate due to electro-mechanical errors and environmental variations
Solution Approach 1:
The patent applies preliminary action by pre-calculating robust drop patterns that account for expected variations in drop volume and placement. The optimization process anticipates potential errors and designs drop patterns that will remain effective even when variations occur, rather than attempting to correct errors after they happen during the dispensing process.
Solution Approach 2:
The patent employs parameter changes by modifying drop pattern parameters (locations, volumes, timing) based on robust optimization that considers variation ranges. The system adjusts these parameters to create patterns that are tolerant of the expected electro-mechanical and environmental variations, improving precision without requiring more complex hardware.
2Manufacturing precision
If traditional uniform drop patterns are used, then the dispensing process is simple, but residual layer uniformity deteriorates due to variations in drop volume and placement
Solution Approach 1:
The patent applies local quality by creating non-uniform drop patterns where different regions receive different drop sizes, volumes, or spacings based on local requirements. The robust optimization tailors the drop pattern to specific areas of the substrate, accounting for local variations in drop placement and volume to achieve uniform residual layers across the entire surface.
Solution Approach 2:
The system performs preliminary optimization of the drop pattern design to anticipate and compensate for variations before dispensing occurs. By pre-calculating the optimal drop pattern that accounts for expected variations, the system achieves uniform residual layers without requiring complex real-time adjustments during the dispensing process.
3Reliability
If drop patterns are optimized for ideal conditions, then theoretical precision is maximized, but actual performance deteriorates when electro-mechanical errors and environmental variations occur
Solution Approach 1:
The patent applies beforehand cushioning by designing drop patterns with built-in tolerance margins that cushion against expected variations in drop volume and placement. The robust optimization creates patterns that maintain effectiveness even when electro-mechanical errors or environmental variations occur, providing a buffer that protects against precision degradation.
Solution Approach 2:
The system changes the approach to parameter optimization by using robust optimization that considers variation ranges rather than ideal conditions. This involves adjusting drop pattern parameters to account for the full range of expected variations, ensuring reliable performance across different operating conditions rather than optimizing for a single ideal scenario.
Data Source
AI summary
Imprint lithography may comprise generating a fluid map, generating a fluid drop pattern, and applying a fluid to a substrate according to the fluid drop pattern. The fluid drop pattern may be generated using a stochastic process such as a Monte Carlo or structured experiment over the expected range of process variability for drop locations and drop volumes. Thus, variability in drop placement, volume, or both may be compensated for, resulting in surface features being substantially filled with the fluid during imprint.