MUNICH -- At ISS Europe 2012 here semiconductor chip, tools and material companies concluded 450-mm wafer production will catapult Europe back on the competitive stage along the U.S. and Asia to produce leading-edge semiconductors sometime in the next 15 years.
The volley has been served.
The EEMI450 is a 450-mm dedicated initiative, which was started in 2009 by European semiconductor equipment and materials industry to bring interested parties together to promote common 450-mm efforts and to induce common 450-mm European projects.
A panel of executives at ISS Europe 2012 debated the consequences of the shift from 300mm to 450-mm wafers.
Alain Astier, Group VP, STMicroelectronics expressed no worries about making the move to 450-mm: “Europe is good and well for 450 mm. The leading 300mm edge today in Europe is both in Crolles, and in GlobalFoundry’s Dresden plants. We are ready.”
Rutger Wijburg, vice president and general manager at GlobalFoundries expressed that 450-mm moves should encompass both scalable More Moore leading-edge chips as well as the less edgy but more system-integrated “More than Moore” ICs: “We need to focus on both.”
Michael Hummel, managing director, Texas Instruments Europe supported the move to 450-mm with a caveat: “In analog, we need to continue to support the 200mm wafer lines even when going to 450-mm.” He hopes the EU will do both.
Hummer emphasized that the move needed to also tackle the cultural and legal aspects, utility costs, and entice young engineers to come into the industry.
Reinhard Ploss, member of the management board and operations, R&D and labor director at Infineon, warned of the need to keep system integration in mind, even when going to scaling the enabling technologies: “If you can understand the customer better you can serve the proper solution; differentiating the end solution is what should drive the need for 450-mm.”
For his part Claus Schmidt, managing director of Robert Bosch Venture Capital urged the EU to continue support VCs: “We need startups with ideas, startups need to go and sell products, otherwise they will vanish.” Cultural differences between entrepreneurs in Silicon Valley and in Europe is evident: startup failures are taken literally in Europe whereas failures by startups in the Valley are chalked up to experience and new ventures are pursued, according to Schmidt.
Paolo Gargini, director of technology at Intel, was confident about the move to 450-mm based on the historical moves from 200mm to 300mm: "It should be smoother this time, we have learned a lot from the past wafer changes. He foresees 1st fab volume production in the 2016-2019 timeframe.
“Traditional scaling took 30 years; from incubation from research to manufacturing is about 10 to 15 years. It’s not a question of when. It’s up to Europe what it wants to do next,” said Gargini. He lauded the decisions made by New York state officials in funding Intel 450-mm plant that is supported by Samsung, TSMC, IBM and GlobalFoundries. “We will be running it as a pilot plant for now. If Europe wants to come in, they are welcome to draft a plan.”
Willy van Puymbroeck, head of unit nanoelectronics, DG Information Society and Media at the European Commission, stated: “The European Commission has received a study carried out by Decision and Future Horizons, one of many, that will be used arrive at a position on 450 mm." He said that “450-mm will arrive with or w/o Europe. It will be much better managed than the transition to 300 mm was. I expect that it will likely be at the 8nm technology node that volume production on 450-mm lines will begin.”
Luc Van den hove, president of research consortium Imec, stated that there is a need to continue to “support 200-mm, 300-mm and absolutely go to 450 for both Moore and More than Moore reasons using European strengths in the equipment and materials.” Imec expects to have a pilot line for 450-mm wafers by 2015. “Europe needs to take leadership role to be there 5 years from now,” said Van den hove.
It defines the goals of the EEMI450 Initiative and emphasizes the importance of early engagement in 450-mm research and development activities for the European semiconductor related equipment and materials industry.
The White paper states that "It will become more economical for chip makers to build one 450-mm factory rather than two 300mm factories."
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.