Review of Evolution and Advances in Photolithography and Nanopatterning Using Free-Electron Lasers

Abstract

Abstract. High-volume manufacturing (HVM) for advanced photolithography is mainly dominated by ASML's scanner technology (Nakamura et al., 2023, “High-Power EUV Free-Electron Laser for Future Lithography,” Jpn. J. Appl. Phys., 62(11), p. 116002). However, the laser-produced plasma (LPP) source used to generate extreme ultraviolet (EUV) light in these ASML tools has significant limitations, including contamination with tin debris, high wall-plug power use, limited control over polarization, and stochastic dose variability (Nakamura et al., 2023, “High-Power EUV Free-Electron Laser for Future Lithography,” Jpn. J. Appl. Phys., 62(11), p. 116002). The main variability, characterized by line-edge roughness, line-width roughness, and patterning defects, arises because the increased energy of the EUV light results in a reduced number of absorbed photons in the photoresist. Due to these limitations, researchers in the semiconductor industry are exploring free-electron lasers (FELs), a promising alternative. FELs produce light by first accelerating electrons through methods such as energy-recovering linacs, compact wakefield accelerators, or storage rings, all of which aim to increase power, tunability, and efficiency. Subsequently, these particles are deployed as radiation sources with varying wavelengths, such as EUV or research-stage water-window X-rays, which can be optically routed to a lithographic exposure tool (Berman et al., 2025, “Ultra-High-Gain Water-Window X-ray Laser Driven by Plasma Photocathode Wakefield Acceleration,” arXiv preprint, arXiv:2507.06403.). The tunability and efficiency achievable through FELs can provide significant advantages in lithography, particularly for fabs manufacturing smaller chip feature sizes with advanced device topologies. While current research on FEL hardware and operation information is limited, early-stage companies have gained traction and attracted significant investment for architectures targeting HVM-class power and uptime metrics, where most data are gathered with simulations and forward-looking claims rather than validated demonstrations. In this review, the evaluation of FEL technologies is performed by comparing their performance metrics with those of the LPP systems, for the purpose of identifying the benefits as well as unresolved challenges in using FEL radiation for photolithography in HVM applications.