Lithographic Substrate Alignment Correction via Dynamic Overlay Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lithographic apparatuses face challenges in aligning substrates with varying numbers or placements of alignment marks, leading to insufficient correction data and potential worsening of alignment due to incorrect input information for process corrections.
Innovation Solution
A method that calculates and applies process corrections by summing weighted overlay variations and residual offset corrections based on previous batches' measured offsets relative to a default model, allowing for flexible alignment adjustments even with different alignment mark types, numbers, or placements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the same number of alignment marks is used for each batch of substrates, then the alignment process is standardized and simple, but the alignment accuracy deteriorates when substrates have different numbers or placements of alignment marks
Solution Approach 1:
The system dynamically adapts the alignment process by detecting the actual number and placement of alignment marks on each substrate batch, then automatically adjusting the alignment model parameters accordingly. This allows the system to maintain high alignment accuracy whether substrates have 16 pairs, 14 pairs, or any other number of marks, eliminating the need for fixed standardized procedures.
Solution Approach 2:
The invention changes the parameters of the alignment model based on the measured alignment marks. When different numbers or placements of marks are detected, the system modifies the mathematical model parameters (such as the number of marks, their positions, and weighting factors) to optimally fit the actual substrate configuration, thereby maintaining precision across varying substrate types.
2Productivity
If process corrections are calculated based on a fixed alignment model, then the correction process is efficient and fast, but the correction accuracy deteriorates when alignment mark configurations vary from the model
Solution Approach 1:
The system performs preliminary measurement of the actual alignment marks on the substrate batch before calculating process corrections. This preliminary action captures the true configuration of marks (number, placement, quality), allowing the alignment model to be adjusted beforehand to match reality, ensuring both speed and accuracy in the subsequent correction calculation.
Solution Approach 2:
The invention implements a feedback mechanism where the measured alignment mark data from actual substrates is fed back into the alignment model. This feedback loop allows the system to compare expected mark configurations with actual measurements, automatically update the model parameters, and recalculate process corrections with high accuracy, maintaining both efficiency and precision.
3Device complexity
If standard alignment marks are used for all substrates, then the measurement process is consistent and simple, but the reliability of alignment deteriorates when marks are missing or misplaced
Solution Approach 1:
The measurement process is made dynamic by automatically detecting the actual presence, number, and placement of alignment marks on each substrate. Rather than assuming a fixed standard configuration, the system adapts its measurement strategy to the actual marks available, selecting appropriate marks for measurement and adjusting the alignment model accordingly, thereby maintaining reliability despite variations in mark quality.
Solution Approach 2:
The system applies local quality by evaluating each alignment mark individually based on its actual characteristics (presence, position, quality metrics). Instead of treating all marks uniformly, the system assigns different weights or selection criteria to different marks based on their local quality, ensuring that only reliable marks are used for critical measurements, thus maintaining high alignment reliability.
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 improves alignment accuracy and reduces overlay errors by enabling precise corrections even when standard alignment marks are missing or misplaced, ensuring consistent and efficient exposure of substrates.
Implementation Method 1
measuring the overlay of alignment marks by reflecting a radiation beam from the superposed marks
Implementation Method 2
the diffracted beam will show no misalignment or overlay of the marks
Data Source
AI summary
Correcting for misalignment of a substrate before it is exposed is performed using offset corrections and process corrections that are calculated based on alignment offset measurements of alignment marks and overlay measurements of overlay targets on substrates in previous batches.