Gradient-based inverse lithography for EUV masks via the waveguide method and a physics-informed neural operator
arXiv:2606.25753v1 Announce Type: new Abstract: Gradient-based inverse lithography technology~(ILT) for extreme ultraviolet~(EUV) masks is presented. A novel framework treats the differentiable waveguide method and the recently proposed waveguide neural operator~(WGNO) as end-to-end physics engines, recovering the permittivity of the absorber of the mask through automatic differentiation of the full forward diffraction model. Numerical experiments on realistic 2D and 3D absorbers of the mask (TaBN, La, U) at $\lambda{=}11.2$~nm show that the considered ILT methods make it possible to obtain a
The continuous drive for higher computational density and advanced chip manufacturing necessitates innovation in lithography, making this research timely for pushing the boundaries of EUV mask production.
Improved inverse lithography techniques are critical for manufacturing advanced chips, directly impacting the performance and cost of AI compute and other high-tech hardware.
The proposed gradient-based inverse lithography utilizing the waveguide method and neural operators offers a path to more precise and efficient EUV mask production, potentially accelerating leading-edge chip development.
- · EUV equipment manufacturers
- · Semiconductor foundries
- · AI hardware developers
- · Advanced materials science
- · Less advanced lithography techniques
- · Companies reliant on less precise manufacturing
More precise EUV masks enable denser and more powerful chips to be manufactured at scale.
This improved chip manufacturing directly contributes to the development of more advanced AI models and applications, creating a positive feedback loop for technological progress.
Nations heavily invested in leading-edge semiconductor manufacturing could gain a strategic advantage in the global technology race, enhancing their economic and potentially military capabilities.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG