Data-driven elemental descriptors for rational design of high-sensitivity extreme ultraviolet photoresists

The rational design of high-sensitivity photoresists for extreme ultraviolet (EUV) lithography is hindered by the lack of quantitative, predictive descriptors that link material composition to EUV absorption. Here, we develop a density-free, data-driven approach that quantifies both elemental sensitivity and impact. We introduce elemental sensitivity descriptors, μ _n and A _n , to estimate how elemental doping alters the linear attenuation coefficient and absorbance without requiring density information of the target material. Furthermore, we define elemental impact Δϕ _i to evaluate whether an element acts as a positive or negative contributor when embedded in high-absorption compounds. We identify I, Te, In, Sn, Sb, Cs, and Bi as top candidates for EUV photoresists, while revealing the detrimental role of H and C as matrix components. This work provides a generalizable strategy for EUV photoresist design.