Projection Exposure Lens Manipulators With Neural Network Wavefront Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional projection exposure apparatuses face challenges in achieving high correction accuracy and update rates for imaging behavior due to imprecise correction commands generated by optimization algorithms, which are either imprecise or require lengthy finite element method simulations.
Innovation Solution
Employing a neural network-based algorithm to ascertain travel commands for manipulators assigned to optical elements in a projection lens, utilizing training data sets to correct wavefront deviations with high precision and speed, enabling improved correction accuracy and update rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If optimization algorithms are used to generate correction commands, then device complexity is reduced, but manufacturing precision deteriorates
Solution Approach 1:
A neural network model is introduced as an intermediary between the wavefront sensor measurements and the manipulator control. The neural network has been pre-trained with simulation data to map wavefront aberrations to optimal manipulator settings, thereby providing high-precision correction commands without requiring complex real-time finite element simulations.
2Manufacturing precision
If finite element method simulations are used to ascertain correction commands, then manufacturing precision is improved, but productivity deteriorates
Solution Approach 1:
The neural network model is pre-trained offline using comprehensive finite element method simulations to establish the relationship between wavefront aberrations and manipulator settings. This preliminary action stores the computational results in the neural network's parameters, enabling rapid inference during actual operation without requiring time-consuming real-time simulations.
3Ease of operation
If conventional optimization algorithms are used, then ease of operation is maintained, but manufacturing precision deteriorates
Solution Approach 1:
The conventional optimization algorithm (mathematical/computational system) is replaced with a trained neural network model that directly maps wavefront measurements to manipulator commands. This substitution maintains operational simplicity as the neural network can be queried in the same way as traditional algorithms, while achieving superior correction accuracy through its training on comprehensive simulation data.
Data Source
AI summary
A microlithographic projection exposure apparatus comprises a projection lens having a plurality of optical elements for imaging mask structures and a plurality of manipulators, each of which is assigned to one of the optical elements and configured to change an optical effect of the assigned optical element by manipulating a property of the optical element along a travel. Furthermore, the projection exposure apparatus comprises a manipulator controller which comprises an algorithm for ascertaining a travel command with travel specifications for the manipulators for correcting a wavefront deviation of the projection lens. The algorithm is configured to ascertain an associated wavefront change of the projection lens from a travel vector with a set of manipulator travels using a neural network and to determine the travel command using an ascertainment result of the neural network from the wavefront deviation.