Unlocking the EUV Frontier: TSMC and NYCU’s Quasi-POM Breakthrough Reshapes Lithography

The semiconductor industry is currently navigating the extreme technical limitations of Extreme Ultraviolet (EUV) lithography as we push past the 3nm node and venture into the Angstrom era. The joint research initiative between National Yang Ming Chiao Tung University (NYCU) and TSMC, focused on 'topological quasi phase-only masks' (quasi-POMs) using molybdenum, represents a pivotal shift in how we approach the mask-making process. By moving away from traditional absorbing structures toward a quasi-phase-only paradigm, this research addresses the persistent challenge of mask-induced image degradation, effectively boosting contrast and reducing the impact of shadowing effects that plague current high-NA EUV implementations. From an industry impact perspective, the transition to molybdenum-based quasi-POMs could be the 'force multiplier' required for the next decade of scaling. Current EUV masks rely heavily on tantalum-based absorbers, which introduce parasitic absorption and height-related shadowing that limit the resolution and process window. By leveraging the optical properties of molybdenum to manipulate the phase of the EUV light rather than just its intensity, the industry can achieve higher resolution without drastically increasing the complexity or power requirements of EUV scanners. This technological refinement allows foundries to achieve tighter pitches and more precise pattern fidelity, which is essential for the complex 3D transistor architectures like Gate-All-Around (GAA) FETs. Supply chain implications are significant, though long-term. Transitioning to a new mask architecture requires updates to the entire photolithography infrastructure, including mask inspection, metrology, and repair tools. Companies like ASML, KLA, and Lasertec will need to adapt their equipment to support the unique phase-sensitive measurements required for quasi-POM technology. Furthermore, the supply chain for high-purity molybdenum targets and thin-film deposition precursors will see increased strategic importance as this technology moves toward commercial production. Looking toward the future, this collaboration signals that TSMC is not merely waiting for external tool improvements but is actively shaping the fundamental physics of the lithography stack. While the technology is currently in the simulation and research phase, the successful implementation of quasi-POMs would solidify TSMC’s leadership in high-NA lithography, providing a substantial competitive moat against challengers in the foundry space. As we approach sub-2nm nodes, these mask innovations will likely become the standard, proving that the future of Moore’s Law lies as much in materials science as it does in hardware architecture.