Engineers working at Stanford University have constructed a radical new computer, based upon a processor that is built around carbon nanotube transistors. The group’s new computer design could even help to sustain Moore’s Law for the coming decades.
Moore’s Law was a phenomenon described by Gordon E. Moore, Intel’s co-founder. Moore produced a paper in 1965, where he noted a doubling in the number of components within integrated circuitry, year after year, since its invention in 1958. Eventually, however, Moore altered his law during 1975 to predict that the number transistors in a die would double, approximately every eighteen months.
Manufacturer’s even started to adopt Moore’s Law to set themselves targets to work towards. Unfortunately, in attempting to double the number of transistors present within integrated circuits, a variety of problems have begun to materialize; heat generation increases, requiring advanced cooling solutions, and current leakage worsens.
During an interview in 2005, Moore himself indicated that the trend would not continue indefinitely. According to TechWorld, Moore explained how we were reaching the limits of what could be logistically achieved:
“In terms of size [of transistor] you can see that we’re approaching the size of atoms which is a fundamental barrier, but it’ll be two or three generations before we get that far – but that’s as far out as we’ve ever been able to see.”
The Turing-Complete Carbon Nanotube Computer
This is where a group of Sanford University researchers step in. To ensure the continuation of Moore’s law, the use of different materials, other than silicon, could be fundamental. It’s thought that carbon nanotube transistors could be more beneficial, as the material is superior to silicon in conducting electricity, and could enable production of faster circuits. H.-S. Philip Wong, one of the team’s research professors, explained the performance advantages:
“CNTs could take us at least an order of magnitude in performance beyond where you can project silicon could take us.”
The research team published their findings in the journal Nature, this Wednesday, showing off a simple computer, harnessing transistors made from carbon nanotubes. Although scientists have long struggled to use these materials, the Stanford Robots Robust Systems Group have finally managed to assemble these nanotube transistors to compile entire electronic circuits. Now, the group has successfully developed an entire computer, built using only 142 carbon nanotube transistors.
Subhasish Mitra, the project’s lead researcher, explained that the use of carbon nanotube transistors has been long discussed as a potential replacement to silicon. He boasts that the team have now put forward exciting proof of the technology’s applicability.
According to the New York Times, the director of physical sciences at IBM’s Thomas J. Watson Research Center, Supratik Guha, explains how impressed she was with the team’s work. She maintains, although the computer is not a performance powerhouse, it represents one of the first steps in the use of carbon nanotube transistors.
The system is described as Turing-complete, meaning it is able to complete any given computation when provided enough time, and was used by the Stanford group to successfully complete a number of basic tasks; for example, the team used their own multi-tasking operating system to perform number-sorting tasks, albeit very slowly.
The current generation of integrated silicon circuits deploy 22 nanometer transistors. Intel, for example, explains how they are attempting to “relentlessly pursue Moore’s Law” using 3-dimensional 22 nm transistors in their range of processors. For reference, a single human hair ranges in diameter from 20 to 200 micrometers.
Industry experts believe that these silicon-based transistors can continue to be scaled down to 5 nm by as early as 2020. However, this could be where the limit is set, and the point at which technology companies would need to turn to alternative materials.
Although Stanford engineers have built a radical carbon nanotube computer, Mitra suggests that this is merely a small part of the team’s overall achievements. They have shown that manufacturers can “… move beyond silicon and its cousins” for use in transistor production.
By: James Fenner