ITRS Update Puts More Emphasis, Faith in Immersion

Every other year, the semiconductor manufacturing industry receives a new edition of the International Technology Roadmap for Semiconductors (ITRS). As this is an off year, the ITRS committee presented last month just an update to last year's edition, but it brings with it some interesting changes nonetheless.

Last year, the new edition of the Lithography chapter presented some significant changes in the Potential Solutions table, such as extending 193 nm lithography (without immersion) to the 90 nm node, and dropping ion projection lithography and proximity X-ray lithography off the chart altogether (see "Potential Tool Solutions Make a Shift ," Semiconductor International , February 2004). This year's update carries that momentum even further, with more significant shifts in potential solutions.

Some of these changes came as a result of the Lithography Technology Working Group defining new criteria for evaluating near-term potential solutions. Namely, "solutions shown for the present and next two nodes must address leading-edge requirements in at least two geographic regions, and all infrastructure including resist and mask must be ready for the timing of the node." Consequently, proximity electron lithography (PEL) was removed from the chart beyond the 45 nm node, and electron projection lithography (EPL) was removed entirely, both because they are solutions for only one geographical region.

Immersion lithography, added as a potential solution last year, continues to move quickly, gaining the confidence of the industry along the way. Listed previously as extending to the 45 nm node, 193 nm immersion lithography is now seen as a potential solution even into the 22 nm node. Meanwhile, 157 nm lithography exists among the possibilities only in conjunction with immersion techniques, and then as a less likely option.

One other area that included noteworthy changes was on the difficult challenges tables. For example, the working group put stronger emphasis on challenges related to immersion lithography. At M50 nm, that includes outgassing and leaching during immersion, and the defect control required as a result of the immersion environment (Table ). At the 45 nm node and below, immersion challenges include developing resists with high refractive indexes, high-index fluids and high-index optical materials to extend immersion lithography to its limits.

Beyond the areas mentioned, few significant changes were made to the ITRS Lithography tables in the 2004 update. By and large, a few numbers were added to fill in various gaps, but little was changed in the coloring or values of the requirements tables. One considerable change, however, was the delay by one year of the data volume requirements for optical masks. For example, what was listed as an expected maximum file size of 324 GB (for uncompressed data for a single layer as presented to a raster write tool) for 2005 is now 216 GB, reaching 324 GB in 2006 instead.

CD control remains a sticky issue, with manufacturable solutions unknown for the stated requirements. As noted in the Lithography chapter highlights, "The U.S. and Japan working groups separately conducted simulation studies that concluded that <4 nm 3σ CD control has no known solutions with any technology presently being developed."