| |
Actually, KLAC is part of that diatribe but it went unmentioned. KLAC for defect detection and process control at the wafer level is necessary for the advancement of the technology. What I did not factor in was the KLAC component relative to the reticle providers such as you mention. I see the market much wider are the wafer level.
Now to the good stuff. Your article and reference is absolutely superb since it proves a point I have made time and time again.
Optical lithography (broad band or g-line) could never be extended below 2.0u (one of many barriers mentioned over the years). This was proven by the physics involved with numerical apertures, lens quality, chemical inadequacies, etal. All that it really meant was that they hadn't figure out how to do it yet. Over the years, lithographers like myself along with countless chemical, materials, and equipment providers figured out how to do it (better materials and higher NA lenses, to name a few) by developing a new bag of "tricks" to keep us from having to go to more complex and costly technologies.
Well, form broad band or g-line lithography tools we moved to the i-line part of the spectrum. I-line is still considered optical lithography in my book. We are now debating whether i-line is valid at 0.25u since we know it is a success at 0.35u technology. So much for optical lithography not being extended below 2.0u or 1.0u for that matter. The bag of "tricks" being used now are improved lenses, step and scan, off axis illumination, Chemical Mechanical Polishing (depth of focus issues), Phase shift masks, improved chemicals (anti reflective coatings), and a host of other things.
Along the way to this recent development technologies such as laser lithography, ion beam, e-beam, and x-ray lithography were to replace optical lithography when it could not longer be used. Every one one of these technologies never saw mainstream implementation at their expected entry points in full wafer manufacturing production. They were delayed as optical lithography marched on for the mainstream. Yes, there are niche pockets of companies using this technology but they never got broad acceptance throughout the industry. These technologies were relegated mostly to the mask making or mask repair sectors of the market as we saw a shift away from the Eaton Mann Optical steppers to the ETEC e-beam systems at the mask shops.
So what does all this have to do with TSI and ArF lasers??? Plenty.
DUV lithography is about to come into its own at the 0.25u or 0.18u device technology. There is a debate as to when it will REALLY be needed and how short its lifetime might be. After all, there are limitations to DUV. It will never be viable below 0.xx micron (pick your favorite number) and other technologies (Scalpel or whatever is being touted now) that will have to be used to allow the migration of device technology below 0.10u for mainstream manufacturing.
WELL WELL WELL, your post seems to indicate otherwise. The work being done by CYMI and its expected introduction of 2 new DUV laser systems seems to indicate the brick wall may be crumbling once again. The article you were kind enough to share shows successful work being done elsewhere. I see in the article they successfully demonstrated the feasibility producing features down to 0.04u using a new bag of "tricks" and improved ArF sources.
Therefore "Science Fiction" may become science fact once again. Now the debate will rage on once more. Is DUV worth investing in if it now looks like it will take you down to 0.04u ??? Hmmm!!!!
Andrew |
|
