LEUVEN, Belgium – European nanoelectronics research institute IMEC sees 2015 as about the time when the first extreme ultra-violet lithography system able to run 450-mm diameter wafers will be available and hence as the time it will need to have a dedicated 450-mm wafer clean room available.
Luc van den Hove, president and CEO of IMEC laid out a timeline to a press gathering here. It began in 2012 with 450-mm wafer tool and metrology testing, moving on to 450-mm process development between about 2013 and 2016 and then into advanced production starting in about 2016.
With the recent announcement of a hub of work for the Global 450 Consortium between five leading chip companies – Intel, IBM, Samsung, TSMC and Globalfoundries – to be based in up-state New York state, it is clear that 450-mm diameter wafer transition is going to happen, said van den Hove.
"We want to continue to do leading-edge, R&D so must make a timely transition to 450-mm, 18-inch wafers in the next few years," he said. He added that getting the timing right is crucial. "There is no sense in doing R&D too early as it would create excessive cost."
Van den Hove said IMEC would take a two-phase approach. The early work up to about 2015 can be done in IMEC's present 300-mm wafer fab which was deliberately made 450-mm compatible in terms of certain specifications such as ceiling height. This work could cover early metrology and process elements, wafer characterizations of stain, uniformity and performance. Phase two would require the ability to hook up machines in a full-flow for process and device development in a production-like environment, he said
Van den Hove said the full-flow phase would require its own clean room and that this would probably be accommodated by a significant extension of the existing pilot wafer fab on the corner of IMEC's Leuven site. He said IMEC was examining various proposals for that modular extension at present.
He also agreed it would be necessary to start building the fab extension prior to 2015 and just in time to receive the first 450-mm wafer capable tools including EUV lithography. Van den Hove said it was hard to imagine that such tools would be available before 2015.
Funding for the extension must be one of the things on van den Hove's mind. The government of Flanders usually supports IMEC on capital projects but as costs increase exponentially along with economic uncertainties, the proportion of funding that IMEC must find from its research fees is almost certain to increase.
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.