In 1965, Gordon Moore, the co-founder of Intel, predicted that the number of transistors per square inch on integrated circuits would double every year and this came to be known as Moore's Law. The silicon industry has succeeded in following this law and transistors have exponentially been decreasing in size since the 1970s. Things are coming to a head, however, and it will be difficult to continue miniaturizing circuits based on top-down, lithographically fabricated bulk silicon transistors.

Nanoelectronics could come into its own here, and researchers are busy looking at developing nanoscale electronic devices and fabrication techniques to produce ever-smaller chips. A nanocomputer, for its part, could overcome the ultimate scaling limitations of conventional semiconductor electronics, says Harvard team leader Charles Lieber. Indeed, a nanoFSM is the perfect model for programmable logic circuits and contains key arithmetic and memory logic elements. An ideal FSM should be able to maintain its internal logic state, modify this state in response to external stimuli and then output commands based on this state to the outside world.

“A basic state transition diagram for a two-bit four-state nanoFSM we looked at in our work (see figure below) shows the four binary representations 00, 01, 10 and 11, and the transition from one state to another is triggered by a binary input signal, 0 or 1,” he explained. “Larger and more complex nanoFSMs can be made using longer binary representations.”

Deterministic nanocombing makes logic tiles

The Harvard researchers fabricated the nanoFSM, according to a design developed at MITRE, using a new technique called “deterministic nanocombing”. They assembled nanowires into six highly ordered crossbar arrays in which each neighbouring pair of nanowire arrays forms a logic "tile". Three logic tiles were programmed so that they performed distinct logic functions and integrated to form a computing architecture. The finished system is able to compute, register data and even update the internal logical state of the nanochip while communicating with the outside world via a digital interface.

“The self-assembled nanowire devices we have demonstrated could be made much smaller than those possible using standard, top-down lithographically defined bulk-silicon transistors,” Lieber told “Until now, it was difficult to organize nanowires at will into dense and efficient nanocircuits, but the new bottom-up technique we used allows us to assemble dense arrays of the many devices required with extreme precision in a pre-defined design,” he said. To date, this could only be done using expensive top-down lithographic manufacturing methods but certainly not with bottom-up assembly.

The technique employed could be a new way to continue miniaturizing a number of electronic systems, he adds. Some examples include biomedical sensors, bio-interface controllers, environmental monitors, drug-delivery vehicles or even insect-sized robots.

According to MITRE's Shamik Das, who was responsible for designing the nanocomputer, another key feature of the nanoFSM is that it requires very little power to run.

The team says that it is now busy perfecting its technology and grafting it onto different materials, sensors and actuators to make more complex nanoelectronics.

The current work is detailed in PNAS doi: 10.1073/pnas.1323818111.

Further reading

Nanolaser tunes in excess of 100 nm (Sep 2008)
Graphene for bioelectronics (Feb 2010)
Tiniest bioprobe breaks new size record (Jan 2014)
Living tissue is laced with electronic sensors (Sep 2012)