Mask Blank Substrate Symmetry Evaluation for Photolithography Flatness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for selecting mask substrates with high flatness for photomasks in semiconductor manufacturing are time-consuming and labor-intensive, and the yield of substrates with high flatness is low due to the variability in substrate deformation when vacuum-chucked, which affects overlay accuracy in photolithography processes.
Innovation Solution
A mask blank substrate is designed with a grid of measurement points on its surface, where differences in height measurements are calculated to ensure that at least 95% of the differences are within a predetermined value, achieving high symmetry and reducing deformation variability, thereby maintaining high flatness and reducing position offsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional substrate selection methods are used, then substrate flatness can be measured, but the selection process is time-consuming and labor-intensive
Solution Approach 1:
The patent applies preliminary action by pre-establishing symmetry evaluation criteria and measurement point patterns on substrate blanks before photomask fabrication. By defining symmetry requirements and measurement protocols in advance, the patent enables rapid assessment of substrate suitability without time-consuming post-fabrication measurements, thus reducing selection time while maintaining flatness precision.
Solution Approach 2:
The patent replaces manual measurement and evaluation processes with an automated symmetry evaluation system. By using defined measurement point grids and computational symmetry analysis, the patent substitutes labor-intensive mechanical measurement methods with automated optical or interferometric measurement systems, significantly reducing selection time while improving consistency.
2Manufacturing precision
If substrates with high initial flatness are selected, then overlay accuracy can be maintained, but substrate deformation occurs when vacuum-chucked
Solution Approach 1:
The patent applies preliminary action by evaluating substrate symmetry and predicting deformation behavior before vacuum chucking. By measuring symmetry at predefined points on the substrate blank and calculating expected deformation patterns, the patent identifies substrates that will maintain flatness stability under vacuum conditions, ensuring overlay accuracy without actual deformation occurring during the process.
Solution Approach 2:
The patent implements feedback by using symmetry measurement results to select substrates that predictably maintain flatness under vacuum. The symmetry evaluation provides feedback about substrate behavior under load, allowing selection of substrates that will not deform excessively when chucked, thus maintaining overlay accuracy through informed selection rather than trial and error.
3Reliability
If symmetry evaluation is implemented, then deformation variability is reduced, but measurement and calculation complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the substrate surface into a grid of predefined measurement points. By evaluating symmetry at discrete, standardized locations rather than continuously across the entire surface, the patent reduces measurement complexity while maintaining reliable deformation prediction. The segmented approach breaks down the complex evaluation into manageable, repeatable measurements at specific points.
Solution Approach 2:
The patent changes parameters by focusing evaluation on symmetry metrics rather than comprehensive flatness mapping. By transforming the measurement approach from evaluating all surface points to calculating symmetry based on selected measurement points, the patent reduces measurement and calculation complexity while improving reliability of deformation consistency prediction through the symmetry parameter.
Data Source
AI summary
In a mask blank substrate having two main surfaces and four end faces, a central point is set on the main surface, a first axis of symmetry that passes through the central point and that is parallel to one of the end faces and a second axis of symmetry that passes through the central point and that is perpendicular to the first axis are respectively set, measurement points are set in the form of a grid with respect to the first and the second axes so as to measure heights of the main surface from a reference plane at the measurement points, respectively, differences each between measured height values at those measurement points located at positions axisymmetric with respect to the first axis are calculated. Those differences corresponding to at least 95% of the total number of the calculated differences between the measured height values are within a predetermined value.