Disruptive R&D

University research spawns startups in everything from PV to resonant detection in MEMS to high-resolution resists for EUV.


Leading university researchers presented their most promising technologies — describing developments ranging from sustainable metal cluster technology (that’s already spawned three notable startups) to resonance-based detection for more accurate MEMS devices — at the new Breakthrough Research Technologies session and the Silicon Innovation Forum at SEMICON West 2014.

OSU metal cluster chemistry spawns startups in resist, displays, and PV
Three interesting startups have emerged so far from the metal cluster chemistry from the Center for Sustainable Materials Chemistry at Oregon State University (OSU), for next-generation photoresists, display backplanes and solar cells. The OSU researchers are developing inorganic materials that can be solution coated in water to form films that match conventional deposition (CVD type) film in quality, and can be designed like polymers for particular properties and have some polymer-like properties. The inorganic metals can then be recovered and recycled from the aqueous solutions.

OSU graphic (Doe)
Source: Center for Sustainable Materials Chemistry, Oregon State University

Inpria is using these metal oxide molecules in its high-resolution resist for EUV and other advanced-node lithography. Replacing the long, tangled, polymer molecules of traditional photoresist with the smaller inorganic molecules enables cleaner edges and reduces collapse of 7nm and 10nm features. CEO Andrew Grenville reported at the Silicon Innovation Forum (SIF) that the line-width roughness with this resist is half that of conventional polymer products (0.7nm vs. 1.5nm) on 10nm lines and spaces. The venture arms of Applied Materials, Intel and Samsung have all invested in the company, and cited it as an example of semiconductor development that they were excited about.

The startup Beet uses metal clusters of Cu3SbS4, a molecule designed to optimize its light absorption to make significantly more efficient solar cells than conventional CdTe or CIGS thin film. “They’re getting 20 percent efficiency with a thin layer, with 31 percent predicted,” said center director Doug Kesler. “We think it has big potential.”

Amorphyx, meanwhile, is commercializing a display backplane technology, capitalizing on a type of electron tunneling through a PE-CVD metal-insulator-metal device for faster switching and potentially significantly lower costs. CEO and president John Brewer said the company is now developing the technology at ITRI in Taiwan with three Asian display makers. VCs selected Amorphyx as one of the best pitch winners at this year’s SIF at SEMICON West.

Georgia Tech drives resonant detection technology for MEMS
On the MEMS side, Georgia Tech continues to develop new sensing mechanisms and devices that use the resonant frequency of vibrating beams to detect chemicals. The beam vibrates at high frequency back and forth in the plane of the device so it doesn’t have to cut through air, so very small changes in mass from chemicals absorbed on the beam’s coating change its frequency. At the Breakthrough Technologies session, Oliver Brand, interim director of the Institute for Electronics and NanoTechnology, reported that the device can detect ~1ppm of toluene without pre-concentration.

Farrokh Ayazi’s group that spawned the Qualtre BAW resonant gyroscope is now working on a new wine-glass shaped gyro and integrating inertial sensors with timing devices, magnetometers and pressure sensors on the same device with the same technology, for 5-10 minutes of reference-free navigation capability. They are also developing encapsulation methods using glass reflowed into etched patterns for glass fill with silicon connections, or sacrificial polymer that is overcoated and then decomposed by the heat of later processing. The group is also developing a low temperature process to build a micromachined ultrasonic capacitive imager on top of a CMOS circuit, so the entire compact unit can be mounted on the tip of a catheter.

University fabs now offer imaging of atomic interfaces and silicon photonics prototyping
University research fabs are more eager than ever to rent out their tools to industry to cover some expense, and have increasingly sophisticated equipment to share. Arizona State University has an electron microscope with a series of non-round lenses and deflectors to correct for the usual spherical aberration, allowing clear imaging of even the usually problematic atomic interfaces. The tool is in a temperature-controlled and electron-field-free room, on a two-meter cement floor to eliminate vibration.

The University of Washington fab focuses mostly on the local biotech sector, and its rapid prototyping service for silicon photonics for the Si-EPIC foundry is growing fast. But Dr. Michael Khbeis, director of the Washington Nanofabrication Facility (WNF), noted that the actual chip processing for these emerging products is actually fairly routine, so he’s always on the lookout for challenging processing problems from outside companies. “We look outside for hard problems with strange materials that can’t run in a normal fab,” he said. “We want to debunk that you can’t trust university tools. We’ve worked hard on ways to prevent cross contamination.”

Maintaining the leading research universities is key to innovation
Universities will likely play an increasing role in generating the research that drives the semiconductor industry, argued Bob Metcalfe, professor of Innovation at the University of Texas at Austin, in his keynote at the Silicon Innovation Forum. He noted that research has also come from corporate labs and government labs, but it’s typically only corporate monopolies that can afford research — and monopolies aren’t usually good for the economy, while government research labs have turned out to not be very efficient. That means the best solution is probably government funding of competing researchers at leading research universities. Innovation also requires entrepreneurs who can scale and take the product from the lab to the market, and venture capitalists who collect money and direct it towards companies that are likely to be successful. “Then it also takes early adopters who are willing to buy the product,” Metcalfe said. “Bless the early adopters. But a problem now is that the economy is so bad that early adopters can’t afford to adopt, so the whole innovation channel is clogged.”

University and research institutions are increasingly important to help accelerate technology advances. The new Breakthrough Research Technologies session and the second SIF at SEMICON West showcased current and future activities, along with the latest innovations and research from collaborative efforts of universities, industries and consortiums. SEMI continues to play a role in using the SEMICON platforms to highlight common issues, emerging trends, and research and enterprise solutions. For more information about SEMI, visit http://www.semi.org and for upcoming SEMICON expositions, visit http://www.semiexpos.org/.