To: Cary Salsberg who wrote (12146 ) 11/19/2004 12:47:01 PM From: Proud_Infidel Respond to of 25522 Litho roadblocks threaten sub-40-nm chip production Mark LaPedus 11/19/2004 12:14 PM EST SAN JOSE, Calif. — New roadblocks are surfacing in next-generation lithography (NGL) that could threaten the IC industry's shift towards sub-40-nm chip manufacturing, warned an official from fab-equipment provider Canon Inc. Current 193-nm optical tools could hit the wall after the 65-nm node, prompting the need for new solutions at 65-nm and beyond. Immersion lithography remains the most promising technology after current and "dry" 193-nm tools, but the jury is still out on whether or not the technology can process wafers -- in volumes. Electron projection lithography (EPL) lacks a customer base, while extreme ultraviolet (EUV) technology has apparently hit a new and unforeseen roadblock -- viable photoresists. EUV is supposed to move into production for the 32-nm node by 2009. The industry must resolve several issues before EUV goes into volume production in fabs, most notably the power source. At a recent EUV event in Japan, lithographers disclosed a new problem. EUV "images below 40-nm with chemically amplified resists are unacceptable," said Phil Ware, a fellow with Canon (Tokyo), a supplier of lithography equipment, during a presentation at International Sematech's Global Economic Symposium on Thursday (Nov. 18). In the past, EUV proponents claimed that current resists were suitable for production. However, Ware indicated that EUV will require "more sensitive resists" in order for EUV tools to process wafers below 40-nm. "It's one of the issues the industry must overcome," Ware said. "EUV looks iffy," he said. "All of the same issues with EUV are still on the list despite all of the money spent in the industry. If EUV doesn't come at the 32-nm node, it will be a challenge at 22-nm." This is not to say that other NGL candidates are a slam dunk, including EPL. "EPL has a different issue: no customers," Ware said. There are even some questions about immersion lithography, which is supposed to emerge in the 2006 and 2007 time frame. While the industry is bullish on immersion, the technology has yet to produce a wafer in a production fab. "What we're waiting for is proof of life," he said. Immersion also appears to have some theoretical limits. Some believe that immersion could scale to the 22-nm node. Ware indicated that an advanced 193-nm immersion tool, coupled with a "flavored water" at a refractive index at 1.6, could only process wafers down to 40-nm. "You still need (a solution) at 30-nm," he said. Immersion with 157-nm tools could go beyond 40-nm, but that technology still suffers from a lack of suitable 157-nm lens materials. And there is no viable liquid for 157-nm immersion, he said.