Gallium oxide electronics withstand extreme cold | Hacker News Reader

Gallium oxide electronics withstand extreme cold

discovery.kaust.edu.sa

I question the claim that electronics are at risk of being too cold in space. It's generally the complete opposite—it's hard to keep electronics from overheating in space, because you can't shed heat by convection and are limited to radiation. And the components all produce heat while operating. It's trivial to insulate things. It's a lot harder to radiate the energy when you want to.

It does sound like this might help with the hot end of the spectrum, though.

I think you are right about the high temperature issues. But extremely cold temperatures also have their problems: the metal contained in electronic devices tends to contract and become more brittle.

From the article:

> At low temperatures, however, electrons become trapped and cannot move, a phenomenon known as freeze-out.

This is a well-known phenomenon. And while you're right that keeping something cool in space requires engineering, keeping things warm in space is also challenging and may require energy expenditure that's otherwise wasteful.

For an artificial example, say you want to put a beacon on an asteroid that emits a signal once a day. You don't want to burn energy keeping the electronics alive or, the beacon won't last very long. You'd like it to sleep without using power, and wake up without needing to warm up. And that means operating at cryogenic temperature.

And of course, chips that work in cryogenic environments are also useful in terrestrial contexts -- we love stuffing refrigerators full of exciting circuitry where I work.

Original paper: https://pubs.acs.org/doi/pdf/10.1021/acs.nanolett.5c06155?re...

The scale is micrometer sizes… I wonder what the limits for shrinking are?

The claim that traditional electronics can't work at deep cryogenic temperatures is vastly exaggerated. Of course you can't take something as complex as a modern microprocessor and hope it works at 4K, but modern fully-depleted MOSFETs typically works. Their electrical characteristic will be different, so a tailored design it's often needed

Exactly. In physics, when people measured things at 4K, there were many examples of active circuits built to measure or amplify things in situ at 4K, and only send the results out, often by the same wire that was used to power the circuit.

The point is not that silicon is not suitable for extreme cold -- it is, with careful design. But the authors want to use a semiconductor which would operate at much higher temperatures than silicon can. And most of these have a disadvantage that they stop working at low temperatures.

What the authors have developed, is transistors which could work at a very high temperature, but also work at extremely low temperature. That is quite rare.

Cool (literally ;)

Now they just need to find something that will work on Venus.

A New Computer Chip Could Finally Withstand The Hellscape of Venus [1]

High-temperature memristors enabled by interfacial engineering [2]

1. https://www.sciencealert.com/a-new-computer-chip-could-final...

2. https://www.science.org/doi/10.1126/science.aeb9934

Recent news :) "operated reliably up to 700 °C"

Meta: the HN code should perhaps be tweaked so that *.sa submissions can handle three-level domains, as the 'base' for anything seems to be two-level (like X.uk):

* https://en.wikipedia.org/wiki/.sa#Second-level_domains

* https://en.wikipedia.org/wiki/.edu_(second-level_domain)

* https://en.wikipedia.org/wiki/.ac_(second-level_domain)

Email hn@ycombinator.com

Is there any irony in cold research coming out of Saudi Arabia?

* https://en.wikipedia.org/wiki/Saudi_Arabia#Geography

Interestingly, the article says that gallium oxide electronics can operate in extreme heat too, up to 500ºC.

Reply on news.ycombinator.com