EUVL at other wavelengths: comparisons and opportunities

Current extreme ultraviolet lithography (EUVL) relies on 13.5nm illumination using Sn-based laser-produced plasma (LPP) sources and Mo/Si multilayer optics, which have enabled high-NA platforms up to 0.55. Further resolution scaling can be linked to a hyper-frequency metric, defined as the maximum spatial frequency resolvable by a coherent projection system, or NA/λ, and allowing for cross-wavelength comparisons using various performance criteria. Three alternative EUV wavelength options are considered and compared to 13.5nm, namely 11.3nm using multilayers of Ru or Mo with Be near the Be absorption edge, 6.6nm using La and B multilayers near the boron absorption edge, and 3.13nm using Sc multilayers combined with Cr or V. For each wavelength scenario, comparative criteria include material properties, depth-of-focus (DOF) characteristics, multilayer penetration performance, spectral and angular bandwidth, reflected amplitude and phase, polarization consequences, and optical scatter. Results highlight that 13.5nm hyper-NA may be one route for next generation EUVL, supported by mature sources and infrastructure, but its feasibility will likely be conditioned by DOF, polarization, and phase error effects. Hyper-frequency EUVL at 11.3nm based on Ru/Be arises as a credible alternative to Mo/Si at 13.5nm, which can provide higher refractive index contrast with sufficiently low absorption, thereby mitigating polarization-driven image degradation with improvements in both angular and spectral bandwidth. DOF and phase error effects can also be significantly reduced, and there is potential to support further NA scaling. At 6.6 nm, La/B multilayers can in principle achieve reflectance above 70% with hundreds of layer pairs and extend hyper-frequency resolution, yet their narrow spectral and angular bandwidths, elevated scatter, and severe layer and interfacial control requirements will likely pose barriers to practical lithography tool integration. Even shorter-wavelength systems in the water window range of 2.3-4.4nm, employing Cr/Sc or other materials, offer the prospect of sub-5 nm resolution, but extremely low reflectivity, high scatter, and immature fabrication processes render them a longer-term possibility rather than a viable solution for near-term EUVL.