Physics-Informed Neural Systems for the Simulation of EUV Electromagnetic Wave Diffraction from a Lithography Mask
2026-03-16 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors explored using special types of neural networks, called physics-informed neural networks (PINNs) and neural operators, to solve how extreme ultraviolet light waves bounce off intricate patterns on tiny lithography masks. They introduced a new hybrid model called Waveguide Neural Operator (WGNO), which speeds up computations by replacing slow parts with a neural network. Testing their methods on known problems showed these networks were both accurate and much faster than traditional solvers, even for new, unseen configurations. Their WGNO model especially stood out for balancing speed and precision, potentially helping design better lithography masks more efficiently.
Physics-informed neural networksNeural operatorsExtreme Ultraviolet (EUV) lithographyWaveguide methodElectromagnetic diffractionNeural network inferenceComputational physicsMask design optimization2D and 3D modelingGeneralization in machine learning
Authors
Vasiliy A. Es'kin, Egor V. Ivanov
Abstract
Physics-informed neural networks (PINNs) and neural operators (NOs) for solving the problem of diffraction of Extreme Ultraviolet (EUV) electromagnetic waves from contemporary lithography masks are presented. A novel hybrid Waveguide Neural Operator (WGNO) is introduced, based on a waveguide method with its most computationally expensive components replaced by a neural network. To evaluate performance, the accuracy and inference time of PINNs and NOs are compared against modern numerical solvers for a series of problems with known exact solutions. The emphasis is placed on investigation of solution accuracy by considered artificial neural systems for 13.5 nm and 11.2 nm wavelengths. Numerical experiments on realistic 2D and 3D masks demonstrate that PINNs and neural operators achieve competitive accuracy and significantly reduced prediction times, with the proposed WGNO architecture reaching state-of-the-art performance. The presented neural operator has pronounced generalizing properties, meaning that for unseen problem parameters it delivers a solution accuracy close to that for parameters seen in the training dataset. These results provide a highly efficient solution for accelerating the design and optimization workflows of next-generation lithography masks.