News

Scientists Take Step Toward Molecular Electronics

• By William Jackson
• March 20, 2008

If advances in computing power and speed are to continue as they have in recent decades, integrated circuits will have to break the size barrier presented by traditional complementary metal-oxide semiconductors (CMOS).

Molecular electronics is one of the possible avenues for advancement. The NIST research is being funded in part by the Defense Advanced Research Projects Agency.

"Molecules are the ultimate in scaling," said Curt Richter, lead of NIST's Nanoelectronic Devices Metrology project and a member of the team that built the tiny resistors. "They are not ready yet, but they are the smallest thing mankind can engineer."

But for the engineering of practical molecular components, scientists would have to work with existing commercial chip materials and manufacturing processes, said Nadine Gergel-Hackett, a research associate at the National Research Center. "If it is to be realized, the first step would be a hybrid" of silicon substrate, typically used in CMOS, coupled with organic molecules.

But the traditional silicon crystal structure used in CMOS is "viewed as not chemically friendly," Richter said. Because of this, little research has been done with layering organic molecules onto them. So the team's first job was to show it was possible to use common silicon substrates.

The team, which includes Gergel-Hackett, Curt Richter, L.J. Richter, C.D. Zangmeister and C.A. Hacker, describe their work in a paper in the April issue of the Journal of the American Chemical Society.

"We demonstrated that you can assemble the molecules just as well" on the "technologically relevant" CMOS substrate as on specially designed chip materials, Curt Richter said.

But is "just as well" good enough? "That's why we fabricated the devices," Gergel-Hackett said. "Our definition of 'good enough' is 'good enough to use in a device.'"

And it worked. The team fabricated two molecular electronic devices, each with a single-molecule layer of carbon chains. They used chains in different lengths for each device. Both demonstrated the ability to resist electronic flow, with the longer chains offering greater resistance, as expected.

The next step is to fabricate a CMOS-molecular hybrid circuit to show that molecular electronic components can work alongside current microelectronic technologies. The research is focused on developing components for high-performance computing and it will still be quite a while before this technology is commercialized, the researchers said.