What struck me at the SPIE Advanced Lithography conference in February was the progress in EUV lithography, which left me more optimistic than last year. After some years of very slow progress with the throughput and source power of the EUV scanners, we are seeing a sudden breakthrough that changed the outlook of many attendants.

TSMC was the source of the optimism. It’s engineers reported they exposed 1,000 wafers in 24 hours on an ASML NXE:3300B equipped with an 80W light source. This is a watershed advance compared to last year, when even reaching a 10W power level seemed challenging.

TSMC has set up a pre-production experiment. Clearly, they have positioned themselves as a pioneer in EUV, with an ambition to insert EUV in N7 high-volume production, and retrofit it into N10 production. Seeing the advances with the power, that seems possible. However, the EUV scanners still need to improve their average uptime.

Looking at the presentations on EUV resists I didn’t see a major breakthrough, only a slow evolution. There were some nice alternative approaches, but these will take more time to mature. It’s becoming clear that some of the resist suppliers had put their development programs on the back burner in previous years. But now, with the progress made with the source power, they have to play catch-up and start coming up with good progress soon.

There was more optimistic news concerning masks, especially around mask inspection and defectivity, as more and more reliable data become available. Imec collaborated on a study which is developing an actinic blank inspection tool (ABI). Using our full field NXE:3100 EUV scanner, we exposed the masks made from the blanks that were inspected with the ABI tool, and correlated the defects in the resulting wafers to the defects found in the blanks. The results show that the new tool can indeed find all the critical blank defects.

Another highlight was the removable pellicle concept. It showed the successful demonstration of a poly-silicon pellicle developed in the previously by ASML is now available for early EUV adopters as a complete pellicle solution with support from ASML.

I didn’t have the opportunity to follow many presentations on direct self-assembly (DSA), but my general impression was that there were no surprises -- progress and momentum remains very solid. The reported defectivity numbers keep going down, as emphasized by Intel which showed a defectivity level that allowed them to state that DSA defectivity can now be removed from the list of potential DSA roadblocks.

As for the alternative approaches, there were several presentations on nano-imprint. The engineers at Canon Nanotechnologies (formerly Molecular Imprints) now officially focus on the flash memory market, where production is somewhat more tolerant of defects. The prototype tool that was presented features sound defectivity and overlay results, mostly champion data that needs to evolve to repeatable specifications. As for the roadmap, Canon is currently focusing their developments to be ready for flash production. Their next target, two years out, is the DRAM market, and another two years out, they want to be in logic manufacturing.

And then there is the perennial promise of e-beam/multi-beam which this year really seems to have stalled. Intel’s Yan Borodovsky admitted as much, saying that e-beam for direct write on wafers is nowhere anymore. Apart from the mask writing business, some people still see a glint of hope for multi-beam technology in very fast prototyping.

ASML has shipped eight NXE:3300B EUV scanners, five of which are exposing wafers at 40W and one at 80W. Its successor, the NXE:3350B, will start shipping later in 2015.

The goal is now to upgrade the sources to a reliable 100W by the end of this year in as many sites as possible in the field. This looks feasible, as ASML has already demonstrated a greater than 100W light source in-house. After that, the target of 250W doesn’t seem all that far-fetched anymore.

All this data together paints a dramatic change of EUV picture compared to a year ago, a change that got people’s attention and that will make some companies reconsider their plans and roadmaps.

By: DocMemory
Copyright © 2023 CST, Inc. All Rights Reserved