The dawn of EUV
The EUV story begins in the mid-1980s in Japan when, building on multilayer mirror research done in Russia in the 70s, Hiroo Kinoshita projected the first EUV images. Labs in the US and the Netherlands soon also began to explore this potential new development in lithography. Originally called ‘soft x-ray’ lithography, the name ‘extreme ultraviolet’ was inspired by the term used by astronomers for the same light wavelengths and photon energies.
In lithography, using shorter light wavelengths enables chipmakers to shrink the size and increase the density of the features (or transistors) on a chip, making the chip faster and more powerful. When ASML was founded in 1984, the industry was using mercury-vapor lamps that produced light of 436 nanometers (nm), known as g-line, and later, ultraviolet (UV) light of 365 nm, called i-line. Early EUV researchers pursued several wavelengths ranging from 4 to 40, but eventually settled on 13.5 as the sweet spot for generating EUV light with tin plasma.
EUV wasn’t the only technology that researchers explored to enable future generations of ‘shrink’. Electron beam lithography and ion beam lithography seemed to be other viable options, but ASML placed an ‘educated bet’ on EUV lithography, because it was the technology that seemed best suited to continue transistor scaling while remaining affordable in mass production.
Not everyone was immediately sold on the idea of EUV technology, however. At a 2020 SPIE conference retrospective, Hiroo Kinoshita, who had then been a researcher at NTT, described the challenges of convincing his fellow scientists that EUV lithography had a chance. “[I presented my results] at the yearly meeting of the Japanese Society of Applied Optics in 1986,” he said. “Unfortunately, the audience was highly skeptical of my talk. However, my belief did not change.”
Andrew Hawryluk, long-time semiconductor industry veteran and then-researcher at the Lawrence Livermore National Laboratory, recalled similar frustrations at the conference. He explained how in December of 1987, a professor came to visit him and his team. After learning about their groundbreaking research in EUV, the professor asked, “But can you actually do anything useful with this stuff?”
“His words haunted me,” said Andrew, who went home for Christmas break and returned two weeks later with a 30-page whitepaper on EUV lithography. He and his team presented the paper at a conference later that year, but even then, finding support was difficult. “You can't imagine the negative reception I got at that presentation,” Andrew recalled. “Everybody in the audience was about to skewer me. I went home with my tail between my legs vowing never to talk about EUV lithography again.”
But a week later, Andrew got a call from William (Bill) Brinkman from Bell Labs, who subsequently succeeded in getting the US Department of Energy to fund national programs for EUV lithography research at Lawrence Livermore and Sandia Laboratories, in what came to be called the ‘Virtual National Laboratories’.
