|
Sematech lithography duo replies to SPIE nanoimprint posting |
|
|
|
Mar 14, 2007 at 08:57 AM |
Within hours of posting one of my SPIE Advanced Lithography week nuggets, “Why doesn’t the industry support nanoimprint?†I had a message from Dan McGowan, Sematech's maven of media relations.
Seems my words had raised a few eyebrows.
When I spoke with Dan, he asked if I had could meet later that day to chat with a couple of the consortium's litho guys to offer their perspectives on my comments. Dan rounded up Mike Lercel and Andrew Grenville, director and associate director of Sematech's litho efforts. As classic jazz sax lines blended with the white noise of steamed-milk machines from the nearby Starbucks, Mike, Andrew, and I talked about nanoimprint litho (NIL).
When I asked Mike if he thought the challenges facing nanoimprint were mainly of an engineering nature, he said that was mostly true, although defectivity and NIL's throughput (or lack of it) were big sticking points. He also questioned whether the process integration issues associated with using NIL for dual-damascene applications would be solvable. But they did agree, more or less, with Grant Wilson's comments that more resources could be put into NIL, also concurring with the opinion that nanoimprint needs a big company to champion it, much like IBM did with CMP back in the 1980s.
I wondered aloud if the guys would consider writing a brief "guest blog" for Chip Shots, and they agreed. Here's what they sent me yesterday.
Although Sematech has not yet launched a major program in nanoimprint, our traditional role has been to support key semiconductor industry infrastructure. So if you look carefully, much of what Sematech does today has cross-cutting applications in many different lithography areas. The semiconductor industry’s primary focus these days is extreme ultraviolet lithography (EUVL) — which, of course, has its adherents and detractors. Sematech has been driving toward solutions for this technology through several efforts, including our Mask Blank Development Center (MBDC) in Albany, NY.
The MBDC recently demonstrated great particle removal at even 30-nm particle size. Although this was targeted toward EUVL, we started a small nanoimprint template cleaning project that builds off of the EUVL substrate cleaning work. Early results from that project also look very encouraging. That is a great example of using well-designed crosscutting infrastructure to support multiple technologies. The mask blank defect inspection equipment is another example of a technology designed for EUVL (to be able to find 30 nm particles), but has applications in reducing defects on optical masks and imprint templates. So while Sematech may not label many of its projects as “nanoimprint,†much of what we do can provide significant benefits to the template-making infrastructure.
It is quite ironic how the “Apollo Program†effects of trying to enable new lithographic technologies end up benefiting competing technologies. For example, the early EUV lens polishing work enabled strong aspheric lens fabrication for 193 nm---and extensions to numerical apertures (NAs) never thought possible. Funding to develop better e-beam mask writers for proximity X-ray litho enabled better optical mask writers capable of handling smaller shapes and more complex optical proximity correction (OPC).
Will the next big “EUVL†breakthrough end up making nanoimprint practical? Or maybe even extend optical litho even further? Difficult to say, but supporting crosscutting efforts, especially in mask fabrication, has usually paid off well.
|
Blogs 
