Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Cymer (CYMI) -- Ignore unavailable to you. Want to Upgrade?

To: Andrew Vance who wrote (15091)	2/26/1998 3:40:00 PM
From: BillyG	Read Replies (1) \| Respond to of 25960

Ultra-deep submicron. It looks like the leading approach for now is 193nm....... techweb.cmp.com Posted: 3:00 p.m. EST, 2/26/98 Researchers plumb the depths of ultradeep-submicron production By Brian Fuller SANTA CLARA, Calif. -- Semiconductor-equipment makers hunkered down this week to tackle the physics problems associated with ultradeep-submicron IC manufacturing. At the annual Society of Photo-Optical Instrumentation Engineers (SPIE) lithography conference here, researchers are touting gains in resists, resolution and line widths and are probing the depths of sub-0.1-micron production, once thought impossible. The advances come as vendors in the semiconductor-gear sector seem less worried about the financial crisis in Asia, where they do much of their business, than about what technology will rule the day below 0.2 micron. "I'd like to see some general agreement on what the next-generation technology is going to be," said Allan Dickinson, vice president of marketing for stepper vendor Nikon Precision Inc. (Belmont, Calif.). "There are five technologies [being suggested], and even SEMI can't figure out what it is. What does the world's leading stepper vendor do? Build all five? That'd be stupid." For now, vendors are trying to stretch optics and are poised to push the wavelength capability of their leading-edge deep-ultraviolet (DUV) systems down from 243 nm to 193 nm; resists are expected to be out late this year. But beyond that, researchers are arguing over the most effective ways to keep Moore's Law on the books. The surprise of the conference was a paper from the University of Texas at Austin in which researchers said they have fabricated a 0.08-micron device using a conventional DUV stepper and special resists. The researchers, led by chemical-engineering professor Grant Willson, used an ISI 10X stepper employing a 193-nm wavelength DUV source. The resist, an amorphous polyolefin, took three years to develop and was made specifically to work with 193-nm DUV. "I didn't believe it could be done at first," Willson said. "It really works better than my wildest imaginings, and it appears that the process latitude is there to generate smaller features yet." Researchers from Hewlett-Packard Co.'s ULSI Research Lab (Palo Alto, Calif.), with an assist from Intel Corp. (Santa Clara), claimed to have tackled the problem of gate-length variations in ultra-deep-submicron design. Because of threshold voltage roll-off and other phenomena associated with short channel lengths, the variations worsen as devices scale down to 0.1 micron and below, disrupting device yield. The team described a six-step "spacer gate" process, using conventional lithography, to control deposition, oxide thickness and etch. "We have produced general patterns that are compatible with integration in a MOS process and have made 100-nm NMOS FETs with 2-nm-thick gate oxide, operating at 1.3 V," the researchers' paper states. The intra-die variation is better than 1 percent (3-sigma). The team also reported having made resistors in silicon films with widths of 50 nm that have shown similar variation. A group from Sandia National Laboratories (Livermore, Calif.) proposed a high-power laser plasma source for EUV systems using a dense beam of large xenon van der Waals gas clusters. The technique is claimed to avoid damage to nearby condenser optical elements when the source cranks out at least 30 W of in-band power. As vendors struggle to expand the life expectancy of optical lithography techniques, IBM is pushing its X-ray efforts. In a paper titled "X-ray fills the Gap," IVM researchers reported having made four critical DRAM levels and a logic level using an SVGL X-ray stepper, claiming overlay results similar to that achieved with optical steppers. Throughput was 61 chips per wafer and six wafers per hour-more than a tenth slower than what conventional steppers can do. Still, the work produced images without the need for proximity corrections and with reduced foreshortening effects and straighter sidewalls. Another novel approach comes out of England. A team from the University of Cambridge suggested replacing the typical patterned transmission X-ray masks with a binary in-line hologram to be projected under near-field conditions. The binary hologram comprises pixels, written at the size of the final reconstruction, that product zero or very small phase changes in transmitted 1-nm-wavelength X-rays. The hologram can be designed to represent any two-dimensional arrangement of conductors or gates. For printing at 50-nm feature sizes, the hologram can be fabricated by electron-beam lithography, in similar fashion to a high-resolution X-ray mask, but in the hologram's case only two phase levels need to be written.