Editor's Page

Focus on the photocell

SPIE Microlithography may be the most important focus event on the semiconductor industry calendar. Nearly 3300 professionals attended this year's gathering held at the Santa Clara Convention Center in late February. Most of the attendees came for the vast symposium program, the leading technical forum in the lithography field. Some of the most interesting papers were presented during two evening poster-session receptions, offering a chance to go one-on-one with authors of the works. The exhibit hall featured only 100 or so companies, but what a potent 100 they were. Most of the key equipment, systems, materials, and component suppliers were there, from the big steppper/scanner and phototrack players to the photoresist makers and metrology vendors. Many exhibitors I queried said the SPIE show was equal to or greater in importance than Semicon West. As one representative noted, "All of our key customers are here."

The leading topics of discussion came as no surprise: how far can optical lithography be pushed and which nonoptical technologies were edging fastest toward commercialization. The halls buzzed with the news that researchers at the University of Texas had successfully printed 0.08-µm features on a silicon wafer using an etched quartz phase-shift mask, a novel amorphous polyolefin resist, and a 193-nm DUV stepper. On the nonoptical front, an executive of one of the major stepper houses said his company was working actively on two of the five "candidate" technologies, with a "foot in the door" in the other three, adding they were waiting to "let the marketplace decide."

The issues of defect reduction and yield enhancement were never far from the surface. Paul Sundland's invited paper at the defect detection and analysis session offered a look at the past, present, and future of defect inspection. He pointed out that because yields change too fast to control, "current inspection systems can't detect certain defects or are too slow." One of his concerns about present technology—and an area for future development—was the lack of "statistically valid measurement of fatal defect levels." Neil Berglund's lunchtime keynote at KLA-Tencor's Yield Management Seminar, which was held in conjunction with the SPIE event, dealt with parametric yield loss in an era of accelerated shrinks. One of his key themes was a differentiation between two major kinds of yield limiters, processing defects and parametric defects. His central conclusion: parametric yield loss management is becoming more and more critical in this age of rapidly increasing device shrinks.

A couple of my favorite quotes from SPIE came near the end of Don Eigler's plenary session talk, "Atoms Where You Want Them." After prognosticating on the emergence of near-atom-scale devices in the next 5 to 20 years (at least in the lab), he pointed out a fundamental limitation on the arbitrariness of atomic manipulation: "You can't get atoms to go where they don't want to go." Eigler then used the term "nanolithography," but joked that there is "no need to turn in your stepper for an STM" (scanning tunneling microscope).

Finally, I encourage readers to investigate our new online version of the magazine, www.micromagazine.com. One part of the site that garners a fair amount of "hits" is our searchable Buyers Guide, which includes "hot links" to most of the supplier companies' own Web sites. We'll be adding features over the course of the year, so any suggestions for improving the site would be greatly appreciated.

Tom Cheyney
Editor

tom.cheyney@cancom.com

MicroHome | Search | Current Issue | MicroArchives
Buyers Guide | Media Kit