UB Technology Solves Chip Fabrication Problem

BUFFALO, N.Y. -- A University at Buffalo professor who has spent the past decade developing a laser ablation apparatus that solves one of the trickier problems in computer-chip fabrication has received a $900,000 grant under the federal Small Business Innovation Research (SBIR) program to commercialize the technology.

A research group headed by James F. Garvey, Ph.D., UB professor of chemistry, has signed a manufacturing-and-marketing partnership with Neocera, Inc., the world's top supplier of pulsed laser deposition equipment, providing the team with automatic access to its market.

With the help of this Phase 2 and other SBIR grants, the UB researchers have established a new local company called AMBP Tech -- short for Assisted Molecular Beam Process Technologies. They anticipate leasing space in the University at Buffalo Technology Incubator this summer.

"We will use the grant and the incubator space to scale up our process to the industry standard and also to ensure our product's compatibility with Neocera's existing commercial systems," said Garvey, principal investigator on the SBIR grant. "Once we do, our laser ablation source then will be added to Neocera's catalog and will be marketed as part of its product line."

In a critical expression of support for the technology, an international leader in computer chips also has signed on as a potential customer and expressed interest in helping to refine the technique.

Garvey, who founded AMBP Tech, is commercializing the technology with his research associate, Robert DeLeon, Ph.D., UB adjunct associate professor of chemistry.

The key advantage of the UB technique -- called Laser Assisted Molecular Beam Deposition (LAMBD) -- lies in its ability to generate thin-film coatings that are produced at temperatures far hotter than the sun's surface and then dramatically cooled to, or even below, room temperature before being deposited onto the surfaces of sensitive electronic devices.

By using such extremely high temperatures and then quenching the heat using a rapid expansion process, the new technique solves one of the more difficult tasks in computer-chip fabrication: how to generate the necessary chemistry to coat the chips without exposing the surface of the device to extremely high temperatures, which often results in costly computer chip failures.

This has been a serious drawback for fabricators of expensive chips for research-grade supercomputers.

"Our process overcomes several hurdles facing companies that wish to generate high-speed chips," explained Garvey. "The use of our technique could enable the fabrication of a whole new generation of devices."

For example, he said, cerium dioxide is an excellent insulator for "silicon-on-insulator" (SOI) applications, which enable high-speed device operation in all kinds of electronic applications.

"Such 'silicon-on-insulator' materials would be an enabling technology for the production of much faster computational devices," Garvey explained.

"This material is just one of the many types of pure metal oxides that the LAMBD source is currently capable of generating."

The LAMBD source also can generate thin films that can be incorporated into such devices as gate dielectrics, which are at the heart of every integrated circuit transistor. In the near future, Garvey explained, the push for smaller devices will necessitate the development of thinner gate dielectrics that are fabricated from advanced materials other than silicon dioxide.

The SBIR grant is being sponsored by the Ballistic Missile Defense Organization and administered by Hanscom Air Force Base.