EUV Lithography: History, Latest Results, Technology Roadmap

In the late 1970s and early 1980s, various research groups - not UC Berkeley, which focused on optical lithography - started to work on next-generation lithography (NGL) technologies. The chief rationale was the “foreseeable” resolution limit of optical lithography being no better than 0.5 µm. In those days, excimer-laser-based deep-ultraviolet lithography was not yet around and high-numerical-aperture (NA) lenses were considered difficult to fabricate without introducing a lot of aberrations. In short, the future of optical lithography looked discouraging. Among the NGL technologies, extreme ultraviolet lithography (EUVL) was actually a later comer, with its long development process starting in the mid-1980s. Never could its pioneers imagine that the technology’s gestation period would be more than thirty years. In the meantime, optical lithography continued to make progress, culminating in 193-nm immersion lithography. Advances in optical lithography carried the geometrical scaling and provided the needed extra years for EUVL to mature.

EUVL entered the high-volume production of semiconductor chips in 2019, at the 7-nm node of logic integrated circuits, and enabled the continued geometrical scaling. Since then, it has been used in research and development as well as the production of advanced logic and DRAM chips, and the performance of EUV exposure systems in both imaging and productivity continues to improve. Meanwhile, after about ten years of development, 0.55 NA exposure systems became available in 2024. The latest lithographic results from such a system will be presented. High-NA EUVL will enable further scaling of logic and DRAM devices, and perhaps new-type devices of the future.

Like optical lithography, the resolution of EUVL, in terms of the half-pitch, follows the well-known expression k1NA. So what’s next? In optical immersion lithography, k1<0.3 is realized in routine production while in EUV lithography the corresponding number is k1=0.4. How to lower the k1-factor further? Do we need an even higher NA? Our presentation will conclude with a discussion on resolution enhancement and ASML’s EUV technology roadmap.

Speakers: Anthony Yen, UC Berkeley, ASML; Ronald Goossens, ASML

Attend in person or click here to watch online