And if EUV fails entirely..?
Defects are also extremely difficult to detect in an EUV system, a situation that's not helped by the fact, Liebmann said, that "the industry as a whole has given up on making defect-free blanks. The plan is to make reasonably defect-free blanks, and then pick a blank to match the chip design, and then align the chip design around the remaining defects so that the defect won't cause any harm."
All these challenges – and more – have been under investigation for quite some time. "We have people within the alliance that have been spending their entire careers on EUV lithography," Liebmann said. "We have some members on the team who have been working on this for literally three decades."
When these researchers explain the problems to him, Liebmann said, "I give them a lot of credibility that when they say there are still some problems remaining, that they actually know what they're talking about."
That said, the plan is now to have EUV up and running at a level that could support high-volume integrated wafer development in 2013, and full-scale deployment at the 14-nanometer process node in or around 2015. "It's going to be extremely challenging to meet that," Liebmann said. "I would say that it's going to be impossible to pull that in from that date."
Liebmann gave a nod to work that is being done on one EUV alternative, maskless massively parallel electronic beam lithography (EBL), citing work being done by Advantest, KLA-Tencor, Vistec, and Mapper. The many varients of EBL essentially eliminate the mask, and use tens or hundreds of thousands of electron beams to do the etching.
"E-beam lithography itself is pretty straightforward," Liebmann said. "We all sort of qualitatively understand how this all works. And we all also understand that the biggest problem is throughput. What all of these systems are working on is massively parallelizing the system to get to a point where you can begin to make this profitable."
From Liebmann's point of view, however, EBL is still not up to snuff. While he's impressed with the companies' prototypes, "None of them are really mature enough at this point even for one of the big exposure-tool companies to pick a solution and run with it."
According to Liebmann, the industry is waiting to see which of the competing EBL technologies is going to be picked up. "Where's Nikon going to go with all of this?" he asked. "Are they going to bet on one of these solutions?"
That said, he and his fellow researchers are keeping their eye on EBL. "We feel that this will be an absolutely vital technology if EUV continues to slip, or if EUV runs into some catastrophic issues, and we have to do the 10 nanometer technology node without EUV."
To stay on schedule for 14nm, Liebmann said, his research group has come up with a transition scheme that doesn't rely on EUV. "We have to do the technology development, we have to do the technology ramp, we may even have to do some early customer demos without relying on EUV, so we'll keep moving forward with 193. To do that, we have to enable some fairly aggressive double patterning or multiple patterning solutions."
To move from double- or triple-mask patterning using 193i to EUV, the 14nm chip layouts will have to be migratable from the optical solutions to the eventual EUV solutions without too much "churn", as Liebmann put it, on the design side. To accomplish that, chip designers will have to keep both optical and EU considerations in mind.
But ultimately, he said, everything should transfer from optical 193i into EUV at the 14nm node. "This is going to be the worst of nodes and the best of nodes, all in one node," he said. "As engineers, it's not going to get boring anytime soon."
After his presentation, Liebmann was asked when he thought EUV would finally be stable, reliable, and have enough throughput for mass production. "Not before 2015," he said. "I shouldn't paint too pessimistic a picture, but you've seen that there's some very severe technical challenges."
The future of EUV is still not certain, he emphasized. "I worked on proximity x-ray lithography for many years," he said. "EUV is still not at the level of maturity where x-ray lithography was when we found out that it's not going to work." ®
The first EUV exposure system, the $65m Alpha Demo Tool (ADT), was installed at the Albany NanoTech Complex at the University of Albany, New York's College of Nanoscale Science & Engineering in 2006. Liebmann quoted IBM Fellow and technology guru Bernie Meyerson as once describing the unwieldy ADT as looking "like someone dragged an electromagnet through a junkyard."
Moore's Law savior EUV faces uncertain future
While it's good to see we're breaking through another technological barrier, part of me wants to see hardware advancement stagnate for a brief period in the hope of inspiring companies to focus on efficient software in order to stand out.
Re: EBL the 3D printer of chip manufacture
"there's always a percentage of bad dies"
Now, hands up who remembers "wafer scale integration" from the 1970s.
Hands up who recognises the name Ivor Catt (stop sniggering, it's the best two British kids TV programs EVER, in one name). If you don't, go reading.
Now, hands up who thinks WSI will be back? Or who has any better ideas?
Why haven't Intel just bought ASML? Presumably Intel don't permit ASML to sell the leading edge stuff to anyone else anyway?
Yep Nikon is a big name, but the biggest is ASML, a Dutch company.
so EUV == X-Ray Projection Lithography. The designation given the wavelength looked odd.
And now I know why. I guess changing the name did not make the technology *any* easier to make work.
I recall Electronics Design articles talking about laser driven metal plasma soft x-ray sources in the 80's. People were saying the load/lock procedures to retain the UHV were a drag on cycle time and the resists exposure times were either *very* long or you needed to put multiple levels on.
I doubt *anyone* thought they would still be *trying* to get it to work about 30 years later.
Anyone still trying with storage rings?