IMEC's MEMS programs seek life beyond Moore's Law

IMEC's MEMS programs seek life beyond Moore's Law
By Tom Cheyney, Small Times Senior Contributing Editor

November 12, 2007 -- Belgium's IMEC has become one of the world's leading R&D centers for advanced semiconductor manufacturing, having built a successful business model around active industrial-partner participation, top-flight team members and program development, and judicious government investment. But there's more to the research group than the relentless pursuit of the Moore's Law CMOS scaling path: MEMS and nanotechnology play an increasingly important role as IMEC moves forward.


The Leuven-based organization uses the term "heterogeneous integration" or the catchy "More than Moore" slogan to describe its programs outside of the advanced chipmaking (or "More Moore") arena. A key component of this multifaceted, multiple-application concept is CMORE, which "opens the 200-mm silicon processing facilities for R&D on silicon-technology-based process steps, process modules and complete processes, targeting the integration of additional functionality or performances surpassing those of standard CMOS processes," according to a recent paper by Lou Hermans, IMEC's NEXT department director and strategic business manager for its new silicon technology applications.

Much of CMORE's activity centers around development of process flows for new MEMS device concept and architectures, Hermans told Small Times. "We are mainly focusing on integrated MEMS, which means a combination of CMOS and MEMS technology. Many years ago, we made a choice to start working on silicon germanium as a structural material. Since then we have been continuously developing that technology. We have a number of projects in place where we are using this technology for realizing or testing certain concepts of certain device structures, [such as] micromirror arrays." Other MEMS applications being explored at IMEC include memory devices using cantilever structures (akin to IBM's Millipede technology), accelerometers, resonators, microbolometers, and thermopile-based energy harvesting.

Hermans cited several reasons for choosing silicon germanium. "It doesn't pose any contamination problems for the CMOS process environment. It is ideal for MEMS because of its high Young's modulus (tensile elasticity), high yield strength, absence of creep, no plastic deformation at typical operation temperatures, and insensitivity to fatigue failure."

IMEC employs what Hermans calls "a MEMS last/post-MEMS" or "monolithic integration" approach to MEMS-CMOS processing. "The MEMS structures are fabricated after completion of the CMOS processing on top of the CMOS wafer. The MEMS structures are realized by pure surface or bulk release etching on the completed CMOS wafer or by the deposition of additional structural layers on top of the CMOS circuitry that afterwards will be structured by surface micromachining."

Since the transfer of IMEC's advanced CMOS scaling activities to its 300-mm development fab, there has been much more access to the 200-mm facility for the heterogeneous integration programs, including MEMS, explained Hermans. "This gives us two things: of course, we have more access to the line, but on the other hand, the line is also becoming more stable. In the past, the scaling people always wanted to have the latest lithography tool and the latest deposition tool, in order to drive the scaling.

"Of course, this is not always what you want to have if you're working on applications where you're really making devices, where you're working toward yield, so you would like to work in a more stable environment. Due to the fact that it is not driven by scaling any more, we end up in an environment that is more stable and more suitable for this kind of work."

Some of IMEC's industrial partners have been showing increased interest in the center's MEMS programs. "There's definitely a growing interest, I think, driven by two things," said Hermans. "We are now in a situation that we can better respond to that interest, but it is also driven to some extent by the fact that some of the companies have stopped or are considering stopping their efforts in scaling in house, or are still doing scaling but only in cooperation with a silicon foundry like TSMC or UMC.

"But they are also looking for new products that they can run in their older fabs. So there's an interest for diversification away from pure digital or analog circuits, to circuits with a higher added value, and MEMS is one of the options."