Researchers have developed a Physics-Informed Neural Operator (PINO) to improve the simulation of electromagnetic scattering problems in extreme ultraviolet (EUV) lithography. This PINO model, which factorizes into lateral and axial branches, is trained using pseudo-spectral frequency-domain (PSFD) equations and learns from approximately 16,000 mask designs without relying on precomputed solutions. The resulting surrogate model achieves a mean absolute error of about $7 imes 10^{-3}$ for scattered intensity on unseen mask patterns, and when combined with spectral damping, it accelerates the PSFD solver for finer discretizations. AI
IMPACT This new PINO model could significantly reduce computational costs and accelerate the design and simulation process for EUV lithography masks, potentially speeding up semiconductor manufacturing.
RANK_REASON The cluster contains a research paper detailing a new model for scientific simulation. [lever_c_demoted from research: ic=1 ai=1.0]
- extreme ultraviolet lithography
- Fourier Neural Operator
- LithoBench
- Physics-Informed Neural Operator
- Pino
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
