C. Grant Willson, professor of chemistry and chemical engineering at The University of Texas at Austin, has won the Japan Prize, an international award similar to the Nobel Prize, for his development of a process that is now used to manufacture nearly all of the microprocessors and memory chips in the world.
He’s sharing the 50 million yen (approximately $560,000 in U.S. dollars) prize with his colleague and friend Jean M.J. Fréchet, who is now vice president for research and professor of chemical science at King Abdullah University in Saudi Arabia. The winners were announced today in a ceremony in Tokyo. The Japan Prize Presentation Ceremony and Banquet, with the emperor of Japan in attendance, will take place in Tokyo on Wednesday, April 24, 2013.
Willson and Fréchet first conceived of “chemically amplifed resists,” the materials for which they are being recognized, in 1979. Willson was a researcher at IBM Corp., and Fréchet was spending a year with the company while on sabbatical from the University of Ottawa.
“My boss came to me and said there is a crazy Frenchman who wants to come and spend a year here. Will you be his host?” said Willson, professor of chemistry and biochemistry in the College of Natural Sciences and the Rashid Engineering Regents Chair in the Cockrell School of Engineering. “I said, ‘Let me do a bit of research on the guy,’ I looked at his papers and they were excellent. He was really doing good science, so I said, ‘Sure.’ He came to join me, and we started by having a great discussion about photoresists.”
Photoresists are light-sensitive materials that enable the basic process used to manufacture computer chips. A pattern of light hits the photoresist, which covers the silicon wafer. The resist becomes soluble at only those points where the light hits. The exposed areas can then be dissolved away to leave patterned access to the silicon. The original pattern can then be transferred into the silicon. In this way the tiny electron-controlling patterns that lie at the foundation of modern computing are produced.
At the time Willson and Fréchet began talking, IBM was the world leader in manufacturing chips. Every two years or so, in keeping with “Moore’s Law,” the company had been able to write smaller patterns on the silicon and thus double the number of devices on each chip. The company was nearing a point, however, when continuing that pace of development did not look possible.
“We were stuck,” said Willson. “Further shrinking of the devices demanded printing with shorter wavelength ultraviolet light. The light bulbs that were available did not produce much light at the shorter wavelength, and the photoresists then being used took hours to develop in response to the low light. It wasn’t practical in terms of production. So we needed to develop new equipment or find photoresist materials that were orders of magnitude more sensitive.”
Willson and Fréchet proposed using a catalyst to amplify the sensitivity of the photoresist. Instead of being dependent on one or multiple photons of light to trigger a chemical change in one molecule of the resist, with catalysts one photon could in theory set off a reaction that would “chew up” many of its neighbors as well. Thus light from the dimmer short wavelength light bulbs would be sufficient.
“It shouldn’t have worked,” said Willson. “It should have been too blunt an instrument to draw fine lines. If you put a cow in a pasture, it will not stay put. It will wander around and keep eating until it eats up the whole field. Our catalysts should have eaten the whole field, but they didn’t. For all practical purposes, they stayed put. We got very high sensitivity and very high resolution. It wasn’t until much later, actually, after the thesis work of two University of Texas graduate students, that we finally figured out why the reaction is controlled in the way it is. At that time of the invention, though, we just needed to know that it worked reproducibly.”
Fréchet, who left IBM at the end of the year, kept collaborating from afar. He and Willson were soon joined by Hiroshi Ito, whom Willson recruited from the State University of New York-Syracuse. Over the next few years the trio developed the process to the point where IBM was willing to put it into production.
“I still remember standing in the clean room at IBM’s facilty in Burlington, Vermont, and watching huge numbers of parts being manufactured with our new material,” said Willson. “It was a thrill that is difficult to describe.”
The chemically amplified resists and their descendants helped IBM maintain its edge in chip production for many years. The patents were licensed in the early 1990s, and many adaptations of the resist were developed. These commercially available materials are now used throughout the industry to enable technologies as diverse as mobile phones, personal computers, home appliances, automobiles and medical equipment.
“The materials have gotten much more sophisticated,” said Willson. “But the fundamental design concept is the same. We made the first cookie, and since then others have made almond cookies and chocolate chip cookies and cookies with a bit of coconut in them that taste better. Hiroshi, who died in 2010, continued to work on chemically amplified resists his entire life and made many important contributions to the modern formulations. If he were alive, he would have shared this prize with us.”
If potential next-generation methods such as extreme ultraviolet (EUV) lithography prove viable, the resists will live on with them. Ironically, Willson himself has placed his bets elsewhere, on a process called nanoimprint lithography that he and S.V. Sreenivasan, a colleague at the Cockrell School, have been developing and commercializing.
“I think that this whole idea of using lasers and lens and resists has reached its limit,” he said. “It’s been amazing, though, to have played the small part in it that we have.”
Source: University of Texas at Austin
