How do you turn a smartphone into a supercomputer? Carbon nanotubes

The future of supercomputing might actually be really, really small.

With support from the National Science Foundation, Stanford’s Subhasish Mitra, associate professor of electrical engineering and of computer science, and H.-S. Philip Wong, professor of electrical engineering, are working with IBM and other collaborators to develop a new generation of computers that have processors based on carbon nanotubes (CNTs).

Roughly 50,000 times thinner than a human hair, CNTs could be the foundation for a new generation of electronics designed to deliver supercomputer performance on something as small and battery-efficient as a smartphone. “Today, we have incredible computing that has changed our lives. But to be able to do more so, we need massive amounts of performance,” Mitra says in the accompanying video produced by the National Science Foundation’s Science Nation online magazine.

Chips based on carbon nanotube designs could eventually complement or replace the current generation of silicon-based electronics.

Today’s silicon chips are two-dimensional, just like a typical suburban landscape of houses and streets. Mitra and Wong are working to create multi-layered chips with carbon nanotube transistors and memory devices stacked on top of each other like an urban skyscraper. This high-rise architecture could boost computing speed and performance into the future.

“Even though it may be the perfect material to make the perfect transistor,” Wong said the main engineering challenge right now is trying to figure out how to make this technology on a large scale.

In 2013, the research team’s work was on the cover of the journal Nature, for progress in building a basic computer using carbon nanotubes. In 2014, the researchers presented a paper at the IEEE International Electron Devices Meeting (IEDM) about how high-rise chips could leapfrog performance of existing circuit cards. Last December, in Rebooting Computing, a special issue of the IEEE Computer journal, the multi-campus team led by Mitra and Wong named this approach N3XT, which stands for Nano-Engineering Computing Systems Technology.

Mitra, whose research focuses on designing robust computer systems and information appliances, is also a member of Stanford Bio-X and the Stanford Neurosciences Institute.

Wong, whose research focuses on topics that include carbon electronics and 2D layered materials, is the founding co-director of the Stanford SystemX Alliance, an industrial affiliate program at Stanford focused on building systems. Before joining Stanford in 2004, Wong spent 16 years at IBM Research working on nanoscale and silicon technologies.

Among other achievements, Mitra and Wong celebrate the success of their former PhD students, including Max Shulaker, who played an integral role in these projects. Shulaker, who recently earned his PhD from Stanford, will continue doing research on nanosystems in a new role that begins in July 2016, when he becomes an assistant professor of electrical engineering at the Massachusetts Institute of Technology.


Substack subscription form sign up