This Chip Could Change Computing Forever

Researchers have created the world's first graphene semiconductor. The joke goes that graphene can do everything but leave the lab, but in the last few years, this is no longer true. In this episode we'll see how scientists turned the best conductor known to man into a semiconductor, opening the door to faster, cooler and more efficient computing.

Note: the resulting graphene was doped with pure oxygen within the experiment. Apologies for not explaining that critical part.

Also another correction, I called "Georgia Institute of Technology" "Georgia Tech" , just wanted to clarify that.

ColdFusion Podcast:

ColdFusion Music:

Get my book:

ColdFusion Socials:

Producer: Dagogo Altraide
Writers: Dagogo Altraide
Editors: Brayden Laffrey
Animator: Tawsif Akkas

[andresconrado]

I've heard, for about 20 years now, that a development just around the corner will make laptops last a week without a charge. Never happens. Color me skeptic.

[am]

"Imagine your phone lasting for days" Why yes, I do remember the 3310

[danbhakta]

When I was young...we used graphite rods encased in wood to do calculations on flat sheets of wood pulp.

[skumancer]

Back in 2003 when I was studying Computer Science at FSU, one of the breakthroughs the university had was regarding graphene. I'm still waiting for it to materialize into ANYTHING we use daily or get a benefit out of.

[sanjitmajumdar]

It's intriguing how often I stumble upon a seemingly groundbreaking technology, only to discover it's been around since the mid-20th century.

[shApYT]

10 years ago: "We gonna have graphene computers!"
10 years from now: "We gonna have graphene computers!"

[new_skyspirit]

Graphene is just like Nuclear fusion, always 10 years away 😅

[Nightenstaff]

Graphene breakthroughs seem to happen just often enough that makes me believe I'll never live long enough to see any graphene products.

[deesh]

The year is 2120... We will finally have graphene computers in 10 years!

[ylpgkxq]

This is the same case as with batteries, every year you hear about some breakthrough tech and still your phone dies in a day with the same old lithium battery.

[AAA-tcuh]

Clicked the video, 40sec in: "Graphene". Rolled my eyes so hard I had to look for them on the carpet.

[snozzmcberry]

At 8:01 you say "deficit free" when the text reads "defect free." The distinction between those two words is VERY important when it comes to semiconductor chips, as wafer defects are a huge deal in the production of semiconductor processing chips.

[me]

A little thing about graphene synthesis. There is a company out of San Diego called Grolltex which has made strides in synthesis and fabrication. Their CEO did his PhD thesis on the subject, too. The main bottleneck is not necessarily the synthesis of pristine graphene, but transferring and fabricating on different surfaces. That's what this company is trying to do.

I know the sad joke about graphene being able to do anything except leave the lab, but the number of companies that are working on scaling, the number of companies working on graphene fabrication infrastructure, and also the very smart people in materials synthesis labs have put out lots of papers recently on the subject. There is one professor at Johns Hopkins who is working on graphene synthesis via CO2 splitting, which is exciting.

If you're an aspiring materials scientist or chemist, this is a great field to be in right now.

[zuu.hed.]

I'm an ingeneer specialized in powerelectronics. We are working since around 2019 with SiC Mosfets.
here are some conclusions in comparison with standard IGBT Moduls (although I obviosly can only talk about powerelectronic side, not microelectronics)
- If they die (explode) they do so pretty quite, that's nice. They also don't push the silicon everywhere. I tell you, cleaning a cabinet where a IGBT did explode is annoying...
- They are faster. witch IGBTs our switchingfrequency was at (depending on the output current) 2...10 kHz, with SiC we are usually at 12...20 kHz with comparable current. That reduces the losses in the chokes significantly, and reduces the noice by far. Sadly it seems, that the chokes are now at their limit. even if the output current of the Semiconductors could be increased, there are no chokes to smoothe the current out. with 25 kHz most choke technologies are also already at the optimum with copper and iron losses.
- They survive overvoltages longer/easier.
- they get extreamly hot, therefore the chip is a lot closer to the baseplate (wich is attached to a heat sink. sadly this increases the parasitic capasity, and because of that also EMC noise against PE. It easily interferes with other devices. An EMC Filter is mostly required.
- because of the higher switching frequency parasitic problems have gone up a notch as well. For example: we build a bidirectional, galvanically isolated DC//DC converter. In the lab it worked with 30 kHz, but when we build it in a cabinet, we had to reduce the switching frequency to 17 kHz, just because the cables were a few centimeters longer, and therefore the parasitic inductance inceased. The difference is, again, the lab compared to a real application.
- they are, at least for now, still very expansive

All in all, it is nice, yes, but it is not the all mighty solution. All the manufacturers have already quited down significantly about innovations.

[ProfessionalGasLighting]

He makes graphene production sound super complex. The Nobel prize winner literally made it by sticking tape together from a pencil.

[nekomakhea]

There's 9 Tech Readiness Levels, and a working graphene transistor is TRL 3. They still have a ways to go.

[ABQSentinel]

I feel the need to point out that, faster clock speeds are not simply limited by the silicon. As clock speeds increase, you start to have a problem where the signals are bIeeding through as RF noise faster than they are traveling across the intended circuit. If the signal "jumps" the circuit gap, then you start to have all sorts of timing problems and the circuit would become increasingly noisy. So, a lot of the circuitry would have to be replaced by opto-electronics which would drive the prices sky-high!

[genjitsu]

I don't know where you got your intro music (you are watching Cold Fusion) and it just really hits the right things. Hard to describe but I really love it so well done! Also I do enjoy your voice and coverage of topics - I do really look forward to your next videos.

[rg]

Graphene based chips? This has been speculated on and teased for many, many years. I’ll believe it when I see it

[escootee]

Very impressive but the most important info is missing: How did they achieve the "semiconductiveness" for graphine? What is going on in detail so the meterial behaves that way?