EDITOR'S PAGE

The litho wars

Bill Oldham is a veteran of the litho wars, so he speaks from experience when it comes to the prognosis for the semiconductor industry's key enabling technology. Since the late 1970s, the UC Berkeley professor noted, the death of optical lithography has been forecast: "Optical lithography's always going to die seven or eight years from now."

Oldham's observations came during his keynote speech, "Lithography Evolution: The Continuing Challenge," at the special plenary session of this year's SPIE Microlithography symposium. Although he had a bully pulpit to express his views, Oldham was but one voice among many. The event brings together the largest concentration of lithographic brainpower from throughout the global chip community, so it's not surprising that expert opinions are as plentiful as unkempt hair among the attendees.

On the next-generation lithography (NGL) front, SVGL's headman John Shamaly says the EUV (extreme ultraviolet) LLC consortium is experiencing growing pains, with more investment dollars needed. His company must concentrate on shorter-term projects, so the big boys (AKA the chipmakers) need to throw something more into the development pot. Shamaly believes EUV will work and will span multiple technology nodes, and that it will be faster than SCALPEL or PREVAIL e-beam projection tools. They're shooting for 80 wafers per hour with 300-mm EUV tools, with 40 wafers per hour easily within reach, the executive notes.

From the SCALPEL camp, Robert Burkhardt, Applied Materials' photomask accounts manager, says the eLith consortium is planning on having its alpha tool ready by the second quarter of 2001, with beta exit scheduled for midyear 2003. He adds that those dates could be pulled in depending on exigencies with either the new phase-shift masks or 157-nm resists. He points the finger at the mask side of the EUV equation, characterizing that part of the food chain as problematic. He also doesn't believe that a robust 157-nm resist can be developed, thus moving the window forward for SCALPEL.

Both Shamaly and Burkhardt see mix-and-match process solutions as part of the evolutionary timeline, with NGL tools performing the critical mask-layer functions alongside advanced DUV optical tools exposing less-critical layers. But some people are floating the idea of mix-and-match NGL for certain types of ICs, with some mask layers done by an EUV ultraviolet tool and others by a projection e-beam-type scanner, with the remaining layers optically exposed. There's little dispute about a mix of one NGL method and optical, but the idea of two next-gen technologies working side by side is intriguing, if a little perplexing, when you consider the logistics of support equipment, facilitization, yield management, and the like.

Each SPIE Microlithography conference has its share of future tech, those seemingly far-out research projects that may never "go commercial." Admittedly, even mainstream NGL proposals have a speculative tinge. One of my favorites comes from the x-ray faction. A megafab-sized "farm" of x-ray litho tools would require several on-site synchrotron sources to make it possible, something most observers see as, shall we say, a showstopper. One can imagine a possible acronym for such an installation -- multiple emission synchrotron system, or MESS.

Plenary headliner Oldham also presented another out-of-the-box idea ­ electron-beam direct write (EBDW) lithography, which is being investigated at Stanford and elsewhere. His description of this maskless litho process and the idea of a massively parallel array of e-beams capable of writing on a variety of materials made me think about Applied Materials' purchase of Etec in a new light. Does Applied's acquisition of the leading e-beam writing-tool supplier mean the mad scientists lurking on those many Santa Clara campuses have something up their collective sleeve? Now that Applied's acquisitive-compulsive condition has spread to the lithographic realm, one wonders how long Morgan, Maydan and their minions will wait to open their corporate wallets to buy an even bigger fish.

Tom Cheyney
Editor

tom.cheyney@cancom.com
http://www.micromagazine.com

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