Smallest Transistor Built with 1-Nanometre Carbon Nanotube Gate


Processor die - Intel

A team of researchers from Berkeley labs has finally broken the scaling limit of 5 nm gate lengths. This breakthrough has set the first foundation stones for the upcoming processors in the future, keeping Moore’s law alive for a little longer.

Why the 5 nm limit?

On the fundamental level, a transistor works by switching the current through the silicon on and off between the source and a drain, responding to the voltage applied to the gate. As we reduce the size of these gates and pack them closer and closer, we tend to approach a quantum problem when the size of the gate approaches five nm. Below this limit, the flow of electrons cannot be stopped. This is due to the quantum mechanical phenomenon known as quantum tunnelling, where the electron simply tunnels from the source to the drain even without the voltage applied to the gate, making it impossible to turn the transistor off. Besides this, the tunnelling also occurs from one gate to the other due to the closed packing, causing havoc.

The 1 nm transistor breakthrough

Smallest transistor built with 1 nanometre carbon nanotube gate

Molybdenum disulphide (MoS2) semiconductor insulated from the single-walled carbon nanotubes (SWCNT) with zirconium dioxide (ZrO2). | Image by Qingxiao Wang, UT Dallas.

This particular transistor is built with 1 nm hollow carbon nanotubes as the gate, zirconium dioxide as the insulator, and molybdenum disulphide as the semiconductor. According to the researchers, this choice of the semiconductor is to avoid the quantum tunnelling effects. As the electrons have higher resistance when flowing through molybdenum disulphide, which also has a lower dielectric constant, they have proved that the flow of electrons can even be controlled at 1 nm size (effective channel length of ~3.9 nm in the off state and ~1 nm in the on state).

While this breakthrough is now a proof of concept, further improvements to make it possible to integrate it into chips for viable computing will be one heck of an achievement.

This post was first published on October 7, 2016.

Geekswipe Team

Geekswipe Team

The official community account for Geekswipe staff members and editors. We are a curious clique of nerds and geeks who co-exist peacefully.


Leave a Reply

Your email address will not be published. Required fields are marked *

6 Responses

  1. Ananthram Sundarababu

    How many Carbon atoms in a 1nm hollow tube?

    • Given that the walls of the hollow tubes are one atom thick (.2 nm) and the tube’s diameter is 1 nm, a single cross-section should obviously have two carbon atoms. But I have no idea how long those CNTs are.

      • Ananthram Sundarababu

        So, the outer diameter will be 1.4 nm at least…. So, that gets us to the question of what are the dimensions of a nanotube?

        • Not sure if that’s accurate. I think the average diameter is close to that. Somewhere between 1.2 nm to 1.4 nm. As for the dimension, CNTs are two-dimensional structures. I am still trying to wrap my head around zero-dimensional stuff. 🙂

          • Ananthram Sundarababu

            Zero dimension? Heard about 21 dimensions of String Theory…. 😊😊😊

            Zero dimension = Just before Big Bang happened, as per modern science? 😎😎😎

            • Zero dimension = Just before Big Bang happened, as per modern science?

              Ah no! :) That’s ‘nothing’ literally! Say, a point! In this context, zero-dimension is something that’s seen in quantum dots where a particle is quantumly confined in all directions so that quantum effects manifest. At the moment, I am still riddled with several thoughts. With enough clarity, I will write up a nifty explanation on Geekswipe soon.