- Binghamton University Research News - https://discovere.binghamton.edu -

Turning down the heat: BU researchers aim to cool microchips

Three Binghamton researchers will work with researchers from GE Research and a New Mexico firm on a $7.2 million, three-year project to solve a problem that has dogged computer chip makers for decades — heat.

For the electronics industry, the BU work may lead to great improvements in how heat generated by microelectronic circuitry is conducted by finding defects in current materials and creating new materials that work better.

Mechanical engineers Bahgat Sammakia and Gary Lehmann and chemist Wayne Jones and the other partners will share in a $3.5 million grant from the Advanced Technology Program of the National Institute of Standards and Technology. GE will contribute $3.7 million to the project. Their task is to develop and test new nanoengineered materials that will yield a 10-fold improvement in the thermal conductivity of materials that play a critical role in the cooling of computer chips. “If we can do that we will really catch people’s attention,” Lehmann said.

BU’s share of the award will amount to $850,000.

The grant is another step in putting BU in the thick of a race to develop new materials and new technology that will make computer chips smaller, faster and more efficient.

“Nanotechnology is a ripe field,” said Jones, who is working with Sammakia on a project funded by Semiconductor International to develop polymer and polymer-metallic hybrids. Also collaborating on that research is Ganesh Subbarayan, a Purdue University mechanical engineering professor who had worked with Sammakia at IBM –Endicott.

The new GE project will further Jones’ and Sammakia’s work on the synthesis and study of the fundamental properties of conductivity in nanomaterials.

Jones will provide the atomic-level chemical analysis and manipulation of materials developed in collaboration with GE and prepared by Superior MicroPowders of Albuquerque, N.M.

Sammakia and Lehmann will concentrate on developing and testing mathematical models that support the theory. Sandeep Tonapi, ’98 ’01, who earned his master’s and doctoral degrees from BU, where he worked with Sammakia, is also a member of the GE research team.

The challenge for the three comes from working with materials defined at the atomic level and measured in nanometers — best described as 1/50,000th of the size of a human hair.

“It’s a real challenge to characterize materials with embedded nanostructures,” said Lehmann who added that performing accurate measurements at this scale “is a leading edge topic in engineering,” both in scholarly circles and in applied commercial engineering circles.

Sammakia said that when modeling the flow of thermal conductivity on a micro or bulk scale — 10-6 — the pattern is constant. On the nano scale — 10-9 or three orders of magnitude smaller — the pattern is more intermittent creating an additional challenge because the models will have to accommodate both variables.

“The size of the particle is key,” he said.

To test the commercial applications, Lehman will develop an apparatus to conduct measurements of heat flow through a 10-100 micron-thick material layer. Math models, based on the nano-scale heat transport, will be used to correlate the measured response with the materials structure.

The work will benefit from the recent acquisition of an atomic force and scanning tunneling microscope as the result of a National Science Foundation grant to Jones and C.J. Zhong in the chemistry department. The devices allow imaging of individual molecules on surfaces such as circuit boards.

The problem to be solved by this project might be illustrated best by comparing the surface of a circuit board to a plaster wall. Even where a wall may appear smooth to the human eye there are ripples and gaps in the surface. On the circuit board these sub-microscopic gaps trap heat generated by layers of diodes and transistors, limiting their effectiveness. Developing a more efficient material to fill in the gaps represents a high-stakes market, estimated at $205 million worldwide in 2000.

Sammakia said creating the ideal material may involve manipulating the shape of the individual atoms to improve their ability to fill the smallest gaps and to improve conductivity.

The two possibilities the research team will study are spherical shapes and cylindrical shapes. The modeling will help determine which shape is more efficient and model structures synthesized in the lab will help verify these observations.

“The geometry of nano-particles can play an important role in conductivity,” Jones said, although exactly what that role is requires a more fundamental understanding of nanostructures and their properties.

Like this article? Please share!