IBM has made significant progress in the search for a potential replacement for today's silicon chips.
A team of eight researchers has discovered how to place carbon nanotubes on a computer chip, and has managed to build carbon nanotube field-effect transistors (CNTFETs) with a density of one billion nanotubes per square centimetre – an arrangement 100 times more concentrated than has ever been managed before.
Carbon nanotubes share silicon's semiconductor nature and CNTFETs can transmit electrons very well when switched on, but have one special advantage. They have the potential to get significantly smaller than silicon chips.
The researchers developed a method through which the cylindrically arranged lattices of carbon atoms can be accurately positioned in individual trenches. More precisely, a series of technical procedures involving advanced chemistry are employed to encourage the nanotubes to arrange themselves.
The team is positive about the breakthrough but is so far refusing to claim that the process will be marketable. The researchers announced, "Carbon nanotubes have the potential in the development of high-speed and power-efficient logic applications. However, for such technologies to be viable, a high density of semiconducting nanotubes must be placed at precise locations on a substrate."
The particularly good news is that the new method is compatible with existing technologies. "This new placement technique is readily implemented, involving common chemicals and processes, and provides a platform for future CNTFET experimental studies," the paper said. "Furthermore, these results show that CNT placement via chemical self-assembly is a promising approach for developing a viable CNT logic technology compatible with existing semiconductor fabrication."
At IDF in September, Intel's senior fellow, Mark Bohr, predicted another decade of Moore's Law but admitted that the 5nm processors it intends to manufacture by the end of that period might not be made of silicon. The company is currently known to be experimenting with the carbon allotrope graphene.
Other potential alternative materials and methods include the involvement of arsenic, gallium, indium, spintronics and silicon photonics.
IBM's study, entitled "High-density integration of carbon nanotubes via chemical self-assembly," is published in yesterday's edition of Nature Nanotechnology.