Lithographic Apparatus Calibration Using Chauvenet Criterion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current calibration and qualification methods for lithographic apparatuses are inefficient, leading to unnecessary rejection of substrate holders and poor repeatability in overlay and focus performance due to the difficulty in accurately measuring parameters like substrate holder flatness and the need for extensive test measurements.
Innovation Solution
A method involving multiple measurements at various positions and times to generate a data set, with a subset selection using predetermined criteria and statistical processes to derive calibration or qualification values, thereby improving the accuracy and reducing errors in substrate holder qualification and apparatus performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional calibration methods using many test measurements are used, then measurement coverage is improved, but measurement precision deteriorates due to poor repeatability and unnecessary rejection of valid substrate holders
Solution Approach 1:
The patent extracts and removes outlier measurements from the data set using statistical analysis (e.g., Chauvenet's criterion). By identifying and eliminating measurements that deviate significantly from the mean, the method isolates the true signal from noise and errors, thereby improving measurement precision while using a large number of test measurements.
Solution Approach 2:
The patent implements an iterative feedback process where measurements are taken, analyzed for outliers, and refined by removing problematic data points. This feedback loop continues until a stable, accurate calibration value is obtained, improving precision through systematic refinement of the measurement data.
2Manufacturing precision
If substrate holder flatness is measured using many test substrates, then measurement completeness is improved, but productivity deteriorates due to time-consuming repeated measurements
Solution Approach 1:
The patent performs preliminary statistical analysis on the measurement data to identify and remove outliers before final calibration calculation. By preparing and filtering the data in advance, the method reduces the need for repeated measurement cycles, thereby improving productivity while maintaining measurement accuracy.
Solution Approach 2:
The patent replaces repeated mechanical measurement cycles with a statistical data processing approach. Instead of physically remeasuring multiple times to ensure accuracy, the method uses statistical algorithms to extract accurate calibration values from a single comprehensive data set, significantly improving productivity.
3Measurement precision
If statistical methods are used to separate substrate holder unflatness from test substrate contributions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces statistical methods (e.g., Chauvenet's criterion, mean and standard deviation calculations) as intermediaries between raw measurements and final calibration values. These statistical tools act as mediators that systematically separate substrate holder unflatness from test substrate variations, improving precision through a structured analytical approach.
Data Source
AI summary
A system and method are provided for qualifying or calibrating lithographic apparatus or parts therefor, using a predetermined objective criterion such as Chauvenet's criterion is used to reject measurement points, individually, by field or by substrate.