An insult sent a weapons scientist home to sketch a new way to make chips

The day before the Christmas shutdown in 1987, Andrew Hawryluk spent the day showing a visiting professor from Cornell around his lab. Hawryluk, a scientist at Lawrence Livermore National Laboratory, and his team had learned how to bounce X-ray light off special layered mirrors, a delicate achievement. The visitor seemed impressed.

Then came the sendoff. The professor looked at him and said, “Can you do anything useful with this stuff?” Hawryluk said he would remember those words to the day he died.

Livermore, set inland in Northern California, was not a chip lab. Founded by Ernest Lawrence, inventor of the Cyclotron, and Edward Teller, a central figure in developing the hydrogen bomb, it was built to design nuclear weapons and had designed more than 10 fusion-type warheads. Part of the work studied conditions inside fusion reactions. Those reactions give off bursts of X-rays, and catching that light offered a way to see inside the reactions and study them.

That parting question stung because the mirrors had no use beyond the lab’s fusion work. So Hawryluk went home furious and spent the next 10 days of the shutdown writing a long white paper. His proposal was to use the mirrors for lithography, the photographic printing step that draws tiny circuits onto silicon chips, but using X-rays instead of ordinary light.

About five months later, he took that idea to a conference. The response was brutal.

“It was extremely negative. That was the low point in my career. I was literally laughed off the stage,” Hawryluk later said. Everyone he admired in the field, he said, lined up at the microphone to explain why it would not work and how misguided the idea was.

He flew home that week. The next Monday he told his boss, “I will never speak of it again.” Three days later the phone rang. The caller was Bill Brinkman from Bell Labs, inviting him to present his work in New Jersey.

Hawryluk walked over to his boss with the news. “Just got this phone call from a guy named Bill Brinkman,” Hawryluk said. “Yeah, he's the Executive Vice President of AT&T,” his boss said, his eyes popping open.

His boss told him in effect he had to go. At Bell Labs, Hawryluk found fellow believers at a moment when Washington was urging the national labs to turn Cold War research into commercial products. Bell Labs teamed up with his lab and a couple of other labs to keep pushing the X-ray printing effort, work that continued to a first international conference in 1993, where the approach was rebranded as extreme ultraviolet lithography, shortened to EUV.

Then in 1996 the government cut funding. Intel, Motorola, AMD plus several other firms pooled 250 million dollars to keep the work going. By 2000 the labs had built the Engineering Test Stand, the first working EUV prototype, which printed features about 70 nanometers wide and showed the idea could work.