By Debra Vogler
WaferNews Technical Editor
At a time when few business opportunities should be ignored, one is the infrastructure required for MEMS manufacturing. This is an area where growth has been limited by a lack of major support in key segments of the equipment industry. Although a number of suppliers recognized the business potential years ago, more work is needed.
“My definition of killer applications are those that can ship in excess of one million units per month,” explains Roger Grace, of Roger Grace Associates.
“Right now, the MEMS market is segmented and splintered with only a few such killer applications, such as air bag accelerometers and disposable medical pressure sensors. The industry needs better processing tools, software, packaging and testing to grow.”
Grace sees equipment suppliers going after the MEMS market ? in fact, they have for at least the past three years. It’s a good thing because he believes that a lack of manufacturing infrastructure has been a barrier to the growth of MEMS as a volume industry.
While a number of equipment suppliers have been on the MEMS market bandwagon, including Karl Suss, Plasma Therm and Electronic Visions Group, Grace doesn’t see the downturn as the catalyst for the trend, except perhaps in the case of Ultratech Stepper. “Ultratech has made a great effort in the last year and is pushing hard to get involved in the MEMS market,” states Grace.
Asked to comment on its MEMS strategy, Fabio Consentino, Ultratech Stepper’s microsystems marketing business manager, described a push for market share that started in the 1988 to 1989 timeframe.
“Throughout the following years, Ultratech Stepper continued its focus on niche applications with medium to high volume requirements and leveraged many of the unique requirements learned from the thin film head industry to directly apply them to other segments of MST/MEMS, such as inkjet print heads, accelerometers, and pressure sensors,” states Consentino. “Today, the thin film head industry is the largest segment of the MST/MEMS industry.”
In a similar vein, Tegal sees the MEMS market as interesting for several reasons articulated by Jim McKibben, VP of worldwide marketing and sales. “We have been focusing on niche applications like MEMS for years,” comments McKibben. “It is a nascent market not served by entrenched incumbents. It is forecast to grow rapidly. Unlike mainstream silicon device fabrication, a MEMS producer requires a variety of profiles depending on the end application for the MEMS device. We intend to be very aggressive in supplying the MEMS market and intend to have platforms capable of meeting the needs of a variety of MEMS producers.”
Semitool has seen the downturn as an opportunity to broaden its efforts in thin film heads, wafer level packaging as well as MEMS. “The MEMS manufacturing processes closely parallel those used in semiconductor manufacturing, therefore it is an ideal fit for our process and equipment technology,” notes Dana Scranton, VP of strategic marketing at Semitool. “We’ve been in the MEMS market for several years … we don’t have to do anything all that different from what we normally do. The processes are adjusted according to the needs of the various devices. Also, many MEMS are integrated with semiconductor devices, so it is a natural fit to market into that segment.”
SUSS Microtec also sees the downturn as a time when pursuing MEMS opportunities is attractive – indeed, as are any technologies that provide for diverse applications. “During the economic downturn, mainstream technology-based companies that manufacture computer or computer-related components might experience hardship in one sector, such as memory or in complementary sectors such as peripherals (e.g., printers),” says Joe Brown, who manages MEMS activities at SUSS Microtec. “MEMS offers attractive market segments where base technologies, such as micromachining, are dispersed broadly throughout industry.”
Brown described a tool that SUSS has developed that allows SOI fabrication from two bonded substrates. “Here is an example where material developed for typical MEMS applications is also used by leading manufacturers of ICs for device fabrication,” notes Brown. “Other products from SUSS developed for mainstream IC fabrication, such as advanced packaging, enable MEMS device fabrication. Examples include thick resist processing, flip chip-bonding and analytical probe test systems.”
DuPont Photomasks has been supplying photomasks for traditional MEMS applications, such as ink jet printer heads and sensors, for a number of years.
“Masks for many of these MEMS applications move through our manufacturing processes very much like masks for ICs, and require minimal to no special handling,” comments Ken Rygler, recently retired executive VP of worldwide marketing and strategic planning at Dupont Photomasks. “MEMS promises to revolutionize nearly every product category by bringing together silicon-based microelectronics with micromachining technology, thereby enabling the realization of complete systems-on-a-chip [not to be confused with MPU, logic and Flash SoCs]. We see opportunity in MEMS as the technology expands into some of the most technologically sophisticated markets, such as optical networking, RF, and biomedical.”
Grace would like to see more companies support the infrastructure needed for “new” MEMS companies.
“Old MEMS companies are focused on mechanical and electromechanical applications,” states Grace. “But RF, optical, bio, and chemical applications are the future of MEMS.” New MEMS companies cited by Grace include Cepheid (microinstrumentation), Iridigm (high-performance displays) and Transparent Optical Systems (optical MEMS).
“Old-guard companies like Motorola and Analog Devices continue to serve the conventional MEMS markets with the companies together cornering approximately 90% of the air bag accelerometer market in the US (approximately 70% globally). However, recently, both companies have entered into the new MEMS markets.”
As for the incentives to enter this market – it’s not only equipment suppliers that stand to benefit. Grace notes that many MEMS manufacturers purchase equipment that had been used for 4-inch and 6-inch wafer manufacturing, so IDMs that might want to sell their outdated equipment can do so.
Using the word “niche” however, seems to be controversial (or at least a matter of semantics). Grace considers the MEMS equipment segment to be a niche area.
“Wafer bonders are unique as are front end/back end aligners,” he explains. “Thick coating is ‘very niche’ and so is deep RIE [DRIE].”
Ultratech Stepper also sees MEMS as a niche market.
“The MST/MEMS market is a significant portion of Ultratech’s business and yes, we do believe there are segments of the MST/MEMS industry with a large potential for growth,” notes Consentino. “However, we call them niche markets because that is how we distinguish ourselves from our competitors. We specialize in niche applications other than traditional semiconductor processing.”
“While some equipment companies may view the MEMS market as a niche market opportunity during this downturn, ASML’s Special Applications Division does not,” explains Norbert Kappel, the division’s director of marketing. “MEMS is an emerging market that is, in a sense, small now, but is likely to grow into a much larger market. It could very well be the next revolution in electronics.”
ASML has been interested in the MEMS market for several years according to Kappel – since 1997 – culminating in the formation of the targeted division in 1998.
An opportunity for equipment suppliers that Grace believes will be very big is in the area of integrated optical MEMS.
“Assembling opto-MEMS requires special tools,” notes Grace. “Now, operators are hand-gluing platforms. I think integrated optical MEMS will be ready for prime time in two to three years.”
Other market segments ripe for the picking are back end MEMS manufacturing equipment, packaging and testing.
“ETEC is the only supplier to the best of my knowledge doing back end MEMS equipment and set up as a merchant market supplier,” says Grace. “Basically, the back end is not very well supported, including scribing, testing, and post-packaging.”
With respect to new killer apps (by Grace’s definition) in the MEMS sector, Grace is cautious.
“MEMS is a disruptive technology, as such, it breaks with the continuum,” states Grace. “I don’t see a killer app in the opto MEMS space but I think there could be killer apps in RF MEMS and in disposable bio and chemical sensors. Essentially, the cost for a MEMS device is broken out into about 75% for packaging and 25% to silicon. Typically, a killer app device would have to cost no more than about $5.”