Chasing After Carbon Nanotube FETs | Hacker News Reader

Chasing After Carbon Nanotube FETs

semiengineering.com

49 pointslnyan10 comments

>The big difference is the channel, which allows electrons to flow from the source to the drain. In today’s transistors, the channel is based on silicon. In contrast, a carbon nanotube FET makes use of a fixed number of tiny and parallel nanotubes for the channels, each measuring 1nm in diameter. Leveraging the properties of these materials, carbon nanotube transistors exhibit high mobilities at low power.

vacuum channel is even better - 460GHz with existing technology https://www.extremetech.com/extreme/185027-the-vacuum-tube-s...

I actually think such vacuum transistors are the future of digital electronics. I remember first seeing the research on them when I was in grad school back in ~2014 or ~2015. The big challenges I see are making a complementary version, to get CMOS-like power savings, and mass producing them with sufficient yield. If you can do both of those, it should be possible to 3D stack them, which would put silicon transistors to shame.

> making a complementary version

How? Electron holes can't traverse a vacuum.

> vacuum channel is even better - 460GHz with existing technology

Interesting. How exactly does this compare? Does anyone know roughly what the switching speed of a finFET is in a leading CPU of today?

I always presumed it to be significantly faster than the cycle speed which is dependent on total propagation of the longest path. Since that path tends to include a series of many switches surely the individual switching speed is significantly faster...

[edit]

Found it, indirectly, lower in the parent's article link:

> In early testing, a vacuum-channel transistor was able to operate at 460GHz — on the order of 10 times faster than a conventional, silicon-based FET (and comparable to the speeds that a graphene transistor might operate at).

Although higher in the article they incorrectly suggest FETs are only capable of a few GHz (probably confusing CPU clock frequency for switching frequency)

along with other serious confusion:

> There’s no word on whether these vacuum-channel transistors will allow for the creation of small, modern audio amplifiers that sound like original tube amps — but maybe!

Switching speed isn’t really what people are after with CNFETs, broadly speaking they’re after parameters like increased drive current per unit of gate capacitance.

Conventional single HEMT FETs already passed 1THz

https://en.m.wikipedia.org/wiki/High-electron-mobility_trans...

I studied FETs, BJTs, and diodes for a year for my engineering degree, so I can mostly read and understand this article.

I don't understand a few things.

Some carbon is put into a furnace with an inert gas, and the output is a bunch of nanotubes. Some conduct like metal, so they can't be used as transistors. Others have a band gap, and these can. I guess it's something to do with the lattice orientation.

Now, imagine that I had a bunch of straws, and a block with straw slots on the top. I can think of no way to randomly disperse the straws over the block to get perfect alignment.

Somehow, the straws end up with good connections to a source and drain, ready for a gate to be wrapped around them.

Deposition of the straws on the substrate appears to be performed both in gaseous and liquid, but the straws must be made elsewhere.

This is completely mistifying.

From 2016, but I guess the general ideas haven't changed:

https://pubs.rsc.org/en/content/articlelanding/2016/nr/c6nr0...

CNTs can be aligned (Fig. 1) via; (i) ex situ, also termed post-synthesis, alignment, where external forces such as electric fields (i.e. dielectrophoresis)4,5 orientate the nanotubes once dispersed onto a substrate, often by means of a liquid medium; or via (ii) in situ (aligned synthesis) where the aligning fields are imposed during nanotube growth.

FET = field-effect transistors

Reply on news.ycombinator.com