Full-flow 450mm line in 2015 says IMEC
Belgian research institute IMEC has stepped up its plans for 450mm development and now aims to have an operational full-flow pilot line by the end of 2015, according to president Luc Van den hove speaking at ISS Europe this week.
At Semicon West last summer, Hans Lebon, vice president of fab and process-step R&D at IMEC, said full-flow validation would probably take until 2016, with individual process modules and wafer handling running up to the end of 2015. In a speech at ISS Europe this week, Van den hove claimed, “In 2015 we will be ready with a 450mm full-flow pilot line.”
According to Van den hove, this would allow early device production on 450mm wafers by the end of 2016 – at roughly the same time as the current schedule for the Global 450 Consortium (G450C) pilot plant being built at Albany, New York. He told TDF that, although IMEC is still working on funding the development of the pilot line, which may include money from the European Union as well as industry, the research institute has started the necessary building work. He said he hoped that the level of European funding would at least match the amount the US and New York State governments have committed to Albany.
Van den hove said the line is intended to co-exist with the Albany-project: “We believe Europe’s role is not to set up a pilot line that is in competition with G450C. It needs to be complementary.”
Van den hove said he sees European-based manufacturing, at the very least at the level of a pilot line, being crucial to the sustained development of the continent’s materials and equipment suppliers.
A key point in the more ambitious timeline being put forward by IMEC appears to be a change in mindset at nearby ASML. The Dutch lithography tool supplier still sees EUV as its priority, according to vice president of marketing Peter Jenkins.
Jenkins spend time in his presentation to caution chipmakers against putting too much faith in the ability of 450mm equipment to cut production costs. “The benefits for lithography are a lot less obvious. The immediate impact of a move to 450mm will be decrease the productivity of the machine and drive up the cost of the process. With previous generations we have used innovations to drive productivity. Those innovations cannot be pushed any further, which is why we have been very hesitant so far to commit openly and support 450mm. But we are ready to support this after we have completed the industralisation of EUV.”
Under ASML’s current timeline, EUV could be in production within two years. However, chipmakers have mostly pushed out their schedules for EUV in favour of double patterning and immersion lithography.
Paulo Gargini, director of technology at Intel, argued that even if the cost of lithography increased or stayed the same with the shift to 450mm, the improvements in efficiency in area-based processes such as deposition, etching and cleaning, which he claimed are still responsible for 60 per cent of the total cost even taking the high cost of lithography tool depreciation into account, would pay for the migration to 450mm.
Van den hove said the cost of supporting 450mm equipment at ASML would most likely be high: “They need to have three tool sets: EUV, immersion and in-air. That is a large development burden but that is where the Europe comes in to help support that. It is why Europe is so critical to this process.”
Gargini said: “I am confident they are investing very aggressively. And I am confident that the cost of lithography will at least come to parity or maybe only slightly worse.”
Several weeks before ISS Europe, Mike Bryant, chief technology officer of Future Horizons, one of the two analyst firms that prepared an as-yet unpublished report for the European Commission on the need for 450mm production in Europe, claimed that ASML was working on a more efficient wafer movement technology that would improve throughput of optical and EUV scanners with the larger wafers.