For several years now, the microprocessor manufacturing world has been coming to grips with the fact that Moore’s law no longer governs the rate at which we are able to shrink transistors in size. This has led to researchers worldwide rushing to find the next big advancement in how microprocessors will be made, and a team of engineers at MIT may have come up with a solution using carbon nanotubes. This is a big deal as carbon nanotubes are much cheaper, environmentally friendly, and faster to produce than the traditional silicon crystals that are used today.
A close up of a modern microprocessor built from carbon nanotube field-effect transistors.
Image: Felice Frankel
To put that in perspective, research has shown that carbon nanotube field-effect transistors, or CNFETs, are up to 10-times as energy efficient as their silicon counterparts, and offer performance gains many times greater than what we can achieve with current silicon-based chips. Unfortunately, at the moment, these gains in speed and efficiency come with the heavy tradeoff of being hard to produce, with lots of defects when manufactured in large quantities.
The team at MIT have worked to solve these trade-offs, and have developed new techniques that greatly reduce the number of defective transistors per run, and to make things even better, the new CNFETs can be produced in traditional silicon chip foundries using the traditional methods used to create the microscopic transistors on silicon wafers. The team demonstrated this by creating a 16-bit microprocessor based on the open-source RISC-V architecture that featured more than 14,000 CNFETs and was capable of performing the same task as similar silicon-based microprocessors.
“This is by far the most advanced chip made from any emerging nanotechnology that is promising for high-performance and energy-efficient computing,” says co-author Max M. Shulaker, the Emanuel E Landsman Career Development Assistant Professor of Electrical Engineering and Computer Science (EECS) and a member of the Microsystems Technology Laboratories. “There are limits to silicon. If we want to continue to have gains in computing, carbon nanotubes represent one of the most promising ways to overcome those limits. [The paper] completely re-invents how we build chips with carbon nanotubes.”
MIT engineers have built a modern microprocessor from carbon nanotube field-effect transistors (pictured), which are seen as faster and greener than silicon transistors. The new approach uses the same fabrication processes used for silicon chips.
Image: Felice Frankel
One of the major hurdles that had to be overcome was intrinsic defects that were present in the carbon nanotubes themselves. Each nanotube will have a percentage of metal in its structure, and that presence of metal causes all kinds of issues when using them to create transistors. To remedy this, the team of researchers developed a technique called DREAM, or “designing resiliency against metallic CNTs.” This technique positions CNFETs in the proper orientation so as not to disrupt the flip of the transistor from on and off. This allows for the use of carbon nanotubes that are just 99.999999-pure while reducing the error rate of bad transistors.
“The ‘DREAM’ pun is very much intended because it’s the dream solution,” Shulaker says. “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.”
With the major cause for failure figured out, the team was able to identify different logic gate combinations that would be robust and not robust with metallic carbon nanotubes. This allowed them to customize the microprocessor’s design that would take advantage of the robust logic gate combinations, and not be affected by the vulnerable connections that could cause failure.
Joining Shulaker on the paper are: first author and postdoc Gage Hills, graduate students Christian Lau, Andrew Wright, Mindy D. Bishop, Tathagata Srimani, Pritpal Kanhaiya, Rebecca Ho, and Aya Amer, all of EECS; Arvind, the Johnson Professor of Computer Science and Engineering and a researcher in the Computer Science and Artificial Intelligence Laboratory; Anantha Chandrakasan, the dean of the School of Engineering and the Vannevar Bush Professor of Electrical Engineering and Computer Science; and Samuel Fuller, Yosi Stein, and Denis Murphy, all of Analog Devices.