Molecular Memory

Replacing silicon with organic molecules could mean tiny supercomputers.

By David Rotman

May 2001

The brains of a computer is the semiconductor chip. Much of the progress over the last 35 years in making computers faster, smaller and cheaper has been a numbers game, squeezing ever more transistors and other electronic devices onto this postage-stamp-sized piece of silicon. Today's PCs pack tens of millions of transistors onto a chip, each transistor as small as a few hundred nanometers (billionths of a meter). But continuing this miniaturization will not be easy-or cheap. Some experts predict that by 2010 it will cost from $30 to $50 billion to build a manufacturing facility to fabricate even smaller chips. Then again, a number of physicists contend the price tag won't matter, because silicon-based devices are fast reaching their fundamental physical size limits.

So what if, instead of carving transistors and other microelectronic devices out of chunks of silicon, you used organic molecules? Even large molecules are only a few nanometers in size; an integrated circuit using molecules could contain trillions of electronic devices-making possible tiny supercomputers or memories with a million times the storage density of today's semiconductor chips (see "Molecular Computing," TR May/June 2000). It may sound like science fiction, but several leading industrial and academic labs are already fashioning rudimentary devices based on "molecular electronics," and several startups have been formed to commercialize the technology.

Nowhere have the advances been more impressive-or the ambitions greater-than at Hewlett-Packard Laboratories in Palo Alto, CA (see "Computing after Silicon," TR September/October 1999). Late last year, HP's research group, led by Philip Kuekes, R. Stanley Williams and University of California, Los Angeles chemist James Heath, received an initial patent on a molecular memory device; a series of related patents covering everything from molecular logic to how to chemically assemble these devices is pending.

"We're trying to reinvent the integrated circuit, with all its functions," says Kuekes, a computer architect. The memory patent, he says, "is fundamental to our strategy. It's one of the legs for us to stand on." Besides being critical to one day building a molecular computer, he says, molecular-based memory incorporated into more conventional silicon microelectronics could be the first commercial application of molecular electronics. Kuekes predicts the group will have a prototype memory device within a year.

The HP patent describes a memory device built from crossbar arrays of nano-wires sandwiching molecules that act as on/off switches. In writing a bit, one voltage decreases the electrical resistance through the molecules, turning the switch "on"; at another voltage, the molecules are turned "off." One advantage of the technology, says Kuekes, is that hardware engineers will be familiar with the design; except for the remarkable fact that it utilizes organic molecules, not silicon or magnetic particles, the technology resembles electronic memories used in PCs. "We want to make sure that what's here is truly useful to build computer systems and is useful in ways that computer designers can understand," says Kuekes.

Story continues below

The HP patent is one of the first to issue that covers a molecular electronic device. However, several groups have patented related nanoelectronics technologies. And a number of other patents on devices that take advantage of molecular electronics are pending at the U.S. patent office.

While it's too early to predict the winner of this race to build and commercialize molecular electronics, those following the field say these initial patents could be critical. "They will be very valuable in all the R&D that builds off them," says James J. Marek Jr., chief executive officer of California Molecular Electronics, a startup that has licensed the rights to several technologies. "These are the building blocks."