I believe in EUVL, I do I do!

Day One of International Workshop on Extreme Ultraviolet Lithography

3 min read
I believe in EUVL, I do I do!

This year’s International Workshop on Extreme Ultraviolet Lithography is at the Sheraton Waikiki, a hotel that overlooks about ten thousand miles of Pacific blue, in which child-sized sea turtles clumsily flap in late afternoon. Earlier today, conference organizer and EUV Litho, Inc. founder Vivek Bakshi caught up with me to remind me that attendees had been instructed to dress informally. I think you know what that’s code for in Hawaii (hint: five meters from the hotel, a store called “Crazy Shirts” makes money hand over fist). And indeed, several Hawaiian-beshirted gentlemen waved to us in flutters of floral reds and yellows and greens while they made their way into the obligatorily-named conference rooms: Kona, Lanai, Maui. Bakshi’s own sartorial decisions had put him about halfway there. He was wearing a Hawaiian shirt in name only. The pattern was recognizable, but muted by the gray, white and black that lent it an almost ironic patina.

One interesting thing about this workshop is the presence of lithography guru and gentleman scientist Chris Mack. Shouldn’t he be at Semicon West, the annual workshop of industry consortium Sematech, underway in Francisco this week, coincidentally overlapping exactly with the EUV Litho workshop?

As it turns out, there’s an interesting little bit of unofficial bad blood between Sematech and EUV Litho. Bakshi, a former Sematech lithography researcher, started EUV Litho, Inc., when Sematech folded their Austin, TX operations and relocated the staff to Albany, New York. Bakshi stayed in temperate Austin and started up his own company, which specializes in workshops just like Sematech does. Apparently the EUV Litho workshop is sort of unofficially boycotted by the folks at Sematech.

But the EUV Litho workshop has acquired the reputation of being more academically oriented where Sematech workshops tend to be focused more on industry vendors and speakers. And according to Mack, this workshop is more interesting. “This isn’t the usual ‘we’re on track, trust us’ talks that the tool vendors always give at similar Sematech meetings,” he says.

Bakshi agrees emphatically. “If anyone gives an industry presentation,” he says, “they are immediately sent out to the poster session.” Consequently, the EUV Litho workshop has developed a reputation for being more academic and R & D-oriented than industry-oriented.

But still, it’s weird to see Chris Mack here. This is a man who notoriously believes that EUV is dying on the vine and that within two years, no one will speak of it again. In fact, earlier this year Mack and Bakshi entered into a bet on the future of EUV lithography in which Mack put up his prized Lotus. Mack’s bet: “Number of submitted abstracts in the area of EUVL for 2011 SPIE Advanced Lithography conference will be zero.”

To hear Mack talk about the future of EUV lithography, you might think you were hearing about a failing political campaign. “A mentality sets in where people don’t want to hear criticisms,” Mack explains. “They think it might sap the momentum. If people doubt the technology, maybe they won’t put the resources into making it work. So the tendency is to gloss over problems.”

“You start to doubt if these talks are really truthful. That’s the bad side of these kinds of workshops. They’re not the really frank discussions of technology gaps that need to be dealt with.”

Or maybe EUV lithography is more like Tinkerbell in Peter Pan Robin Hood *. If you start stop believing in her, she dies.

Mack is not the only one with grave doubts about EUV. At SPIE this past March, IBM’s Bernie Meyerson, ever one to turn a colorful phrase, remarked that ASML’s Albany EUV alpha-demo tool “looks like you rolled an electromagnet through an automotive junkyard.”

Meyerson also said that “scaling is dead. Moore’s Law isn’t.” This led Mack to speculate that Meyerson was signposting that EUV would probably die in the water as more companies looked for alternatives to scaling to make their chips run better, faster and at lower power.

But Bakshi is confident that EUV lithography will prevail. “This is a multi-node technology,” he says. “It will take you to the end of Moore’s law, and there will be no need for any more expensive optical tricks.”

Semicon West
and EUV Litho are both dealing with three main issues that stand in the way of EUV lthography being as ubiquitous as 193-nm litho is today. Those issues are metrology, contamination, and bright light sources.

On Monday, light source manufacturer Cymer Inc. reported that that it has shipped a laser-plasma EUV lithography source that was capable of 75 Watts of exposure power.The company thinks it can achieve wafer production of 60 silicon wafers per hour, the minimum requirement for EUV lithography to be adopted industry-wide. Bakshi said "everyone is thriled something is shipping." But he would not comment on whether 75 Watts was a realistic number.

*Wow, that jet lag is really something, isn't it.  

The Conversation (0)

Two Startups Are Bringing Fiber to the Processor

Avicena’s blue microLEDs are the dark horse in a race with Ayar Labs’ laser-based system

5 min read
Diffuse blue light shines from a patterned surface through a ring. A blue cable leads away from it.

Avicena’s microLED chiplets could one day link all the CPUs in a computer cluster together.


If a CPU in Seoul sends a byte of data to a processor in Prague, the information covers most of the distance as light, zipping along with no resistance. But put both those processors on the same motherboard, and they’ll need to communicate over energy-sapping copper, which slow the communication speeds possible within computers. Two Silicon Valley startups, Avicena and Ayar Labs, are doing something about that longstanding limit. If they succeed in their attempts to finally bring optical fiber all the way to the processor, it might not just accelerate computing—it might also remake it.

Both companies are developing fiber-connected chiplets, small chips meant to share a high-bandwidth connection with CPUs and other data-hungry silicon in a shared package. They are each ramping up production in 2023, though it may be a couple of years before we see a computer on the market with either product.

Keep Reading ↓Show less