Carbon nanotube processor breakthrough made at MIT

Carbon nanotube processor breakthrough made at MIT

Technology News |
By Peter Clarke

The work, which was supported by Analog Devices Inc., the National Science Foundation, and the Air Force Research Laboratory, was reported in Nature.

Carbon is being investigated as the active material in leading-edge transistors as it offers a faster alternative but one advantage of the research demonstrator made at MIT is that beyond the specialized carbon nanotube FETs (CNFETs) the development used standard commercial design tools and manufacturing infrastructure.

Silicon transistors have served the electronics industry well for many decades but scaling of the speed and physical size of these transistors has become increasingly complex and expensive; driving research into alternative materials and physical orientations.

Research indicates CNFETs could offer 10x the energy efficiency and 3x higher speeds than silicon but defects introduced during the, as yet, unscaled manufacturing process has hindered adoption.

MIT’s major contribution is to have invented techniques fabricate that limit defects and enable controllable complementary CNFETs to be made and which can then be used as surrogates for CMOS manufacturing in conventional design flow.

The team used platinum or titanium metal contacts to define the transistors as P- or N-type and then coated them in an oxide compound them to tune the transistors for performance or power consumption.

MIT’s solutions include RINSE (removal of incubated nanotubes through selective exfoliation), MIXED (metal interface engineering crossed with electrostatic doping) and DREAM (designing resiliency against metallic CNTs).

The Nature paper describes the microprocessor design and includes more than 70 pages detailing the manufacturing methodology.

Next: “Hello World!”

The result is a 16bit microprocessor built to the RISC-V instruction set architecture using more than 14,000 CNFETs. As a proof of its functionality the research team were able to program it to send the message: “Hello, World! I am RV16XNano, made from CNTs.”

Much of the research work carried out by the team was designed to find ways round metallic CNTs that had either mandated unfeasibly high purity CNTs or killed transistor operation. By a series of work-arounds the team has allowed off-the-shelf materials to be married with conventional design flows and get CNT into practical production.

“The ‘DREAM’ pun is very much intended, because it’s the dream solution,” said Max Shulaker, the MIT physicist who led the work, in a statement on the MIT website. “This allows us to buy carbon nanotubes off the shelf, drop them onto a wafer, and just build our circuit like normal, without doing anything else special.”

The researchers have started implementing their manufacturing techniques in a commercial foundry – believed to be Skywater Technology Foundry Inc. (Bloomington, Minnesota) –  through a program run by the Defense Advanced Research Projects Agency. Shulaker is reported as saying improved versions of CNT chips could be in the commercial market place within five years.

Related links and articles:

News articles:

3D IC combines CNT transistors, ReRAMs and sensors

Nantero: Memory: The Next Big Opportunity for Differentiated Mobile Devices

SkyWater partners for carbon nanotube design service

CEO interview: Sonderman on SkyWater’s technology foundry model

Carbon nanotubes cause cancer in same way as asbestos

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles