Fusion Progress: Superheated Gas Kept Stable For 5 Milliseconds

An anonymous reader writes: A company called Tri Alpha has successfully kept a ball of superheated gas stable for a record time, 5 milliseconds, putting them closer to producing fusion power. "'They've succeeded finally in achieving a lifetime limited only by the power available to the system,' says particle physicist Burton Richter of Stanford University in Palo Alto, California, who sits on a board of advisers to Tri Alpha. If the company's scientists can scale the technique up to longer times and higher temperatures, they will reach a stage at which atomic nuclei in the gas collide forcefully enough to fuse together, releasing energy.

Importantly, the Tri Alpha machine may be able to operate with a different fuel than most other fusion reactors. This fuel-a mix of hydrogen and boron-is harder to react, but Tri Alpha researchers say it avoids many of the problems likely to confront conventional fusion power plants." The article does not say how much this success cost the privately-funded Tri Alpha, but it certainly wasn't in the billions of dollars.

Comments

24/7 here we come...

[bhlowe]

Expect 24/7 operation in 30 years.

Re:24/7 here we come...

[Anonymous Coward]

As amusing as this kind of comment is, we need to remember that it's easy to be the cynic in the peanut gallery. We're not the ones doing the work while being laughed-at, nor usually even the ones willing to invest in it. From that perspective we don't really deserve to see it in our lifetimes.

Re:24/7 here we come...

[Sechr Nibw]

Well, I was going to point out that you were probably referring to https://xkcd.com/678/, but you got the reference wrong, if that's what you were shooting for. The correct reference is "It has not been conclusively proven impossible".

Re:24/7 here we come...

[pushing-robot]

35, actually. Everyone knows fusion power plants become available in 2050.

Re:24/7 here we come...

[AndyKron]

We're also not the professionals that keep telling us it's 30 years away.

Re:24/7 here we come...

[TeknoHog]

Well you can have my cold fusion when you pry it from my glowing plasma hands!

Re:24/7 here we come...

[rch7]

ITER is expected to be finished in 2019 and start full experiments in 2027. It isn't expected to maintain fusion pulse for more than few minutes or generate any electricity. Once/if they make real research progress, next stage may be DEMO, it may be 2 GW plant, but is not expected to be commercial and is basically vaporware. First commercial station will be after DEMO some time too far in the future.

And in between, we already have very stable fusion plant up in the sky every day. It is called "Sun" and provides more than enough of cheap energy right now, and it gets cheaper each year. I doubt that bringing part of it to Earth would make things easier or reduce your electricity bill, rather the opposite. Though of course it is important for science progress, maybe some space travel and similar things.

Re:24/7 here we come...

[Applehu Akbar]

"avoids many of the problems likely to confront conventional fusion power plants.""

Where is this magic industrial park when I can drive down a row of conventional fusion plants?

record ?

[mZHg]

I don't understand the difference between this record compare to current record of holding plasma, which is about 16min.
https://en.wikipedia.org/wiki/...

Re:record ?

[Crashmarik]

http://news.sciencemag.org/phy...

There's a link explaining the differences.

That's a pretty big if....

[mark-t]

"If the company's scientists can scale the technique up to longer times and higher temperatures...."

Not when... but if.

So, basically, not in anyone's lifetime that is alive right now.

Start expecting it in five.

[Ungrounded Lightning]

We have been expecting cold fusion in 30 years for about 50 years now.

Actually it's HOT fusion we've been expecting in 30 years for a long time. (Cold fusion, other than the apparently useless muon-catalyzed form, was a "maybe it's possible - no apparently not" flash in the pan)

But THIS one is big: It's not that it lasted 5 ms. It's that it lasted 5 ms WITHOUT DECAYING. That almost certainly means that:
  - either they've completely solved the instability issues and it's just a matter of scaling up (and using superconductors or adequate cooling so they can run continuously),
  - or they've solved them well enough to hold the plasma ball together until it's paid for itself several times over, then make another one (repeat continuously) and it's just a matter of scaling up (and using superconductors or adequate cooling so they can putt-putt-putt continuously).

Now if other problem show up (but aren't a fundamental refutation of this indication of stability) we might end up expecting fusion in five years for another fifteen or so. But I think the "30 years forever" thing has just been evicted from fusion and is living with its brother in copyright extension.

Re:Long way to go

[Beck_Neard]

No, it will just destroy the equipment from the intense heat and radiation flux.

The point of aneutronic fusion isn't lifelong operation. You pretty much forfeit any chance of that when you deal with the power fluxes necessary for fusion. The point of aneutronic fusion is elimination or reduction of high-level radioactive wastes which are extremely dangerous and take centuries to decay.

Not that far when you think "voltage"

[Ungrounded Lightning]

to do what they want means they need 3 billion degrees to ignite and they are at 10 million

Each electronvolt is equivalent to 11,500 degrees Kelvin. So they need to run at about 200 kV instead of 870V. Piece of cake.

This is whyFarnsworth fusors are tabletop "gassy vacuum tubes" and the issues with polywell machines are things like geometry and electromagnet wiring rather than applying excitation energy.

Kelvin is the same size degree as celsius but offset by a couple hundred degrees so zero is absolute zero. At 3 billion degrees the difference between water freezing and absolute zero is noise. If TFA's degrees are fahrenheit the offset is still noise but scale the voltage back to 144 kV.

Re:Not that far when you think "voltage"

[Ungrounded Lightning]

First off, there is a big difference between something like a fusor which is basically accelerating a beam of particles to some amount of eV that is similar to the applied voltage, and something going for a thermal distribution with same amount of eV spread out with a tail of the distribution that does most of the reactions

Fusors and polywells aren't about beams. They're about assembling a plasma object that is already hot, by compressing it during the assembly.

The fusor does this by having two concentric spherical electrodes, the inner one skeletal, with a large voltage between them. Positive ions fall inward essentially radially, accelerated by the field until they pass through the inner electrode, and fly on orbits that pass through the center of the spheres. They "pile up" as they pass through the center, thus mapping the acceleration voltage directly into compression temperature as well as high average density. (Unfortunately a small number of ions hit the inner electrode on each pass and are lost. So though it's a great fusion-neutron source breakeven isn't in the cards.)

The polywell does the same thing to electrons - with the added tweak that the inner electrode contains a set of magnet coils that get the electrons to travel in paths that mostly miss the electrode. As they orbit through the center the high average density there is effectively a third high-voltage negative electrode, producing a radial electric field between this "virtual electrode" at the center and the inner physical electrode. Positive ions fall in toward the virtual electrode (nearly neutralizing it) and again you get a high density and inward velocity, mapping the electric field into temperature.

It looks to me like the field-reversed configuration does the same sort of thing, compressing the plasma in a way that maps the electric fields (both directly applied and created by the magnetic field change) into particle acceleration during the compression, and thus into temperature. Unlike Tokamaks and similar devices, you don't "put a low-density plasma in a (magnetic) can" and then have to heat it up. You heat it by squeezing it when you initially assemble it, accelerating the particles toward each other, and that maps your compression forces into temperature - which turns a moderately high voltage into a relative particle speed that has a hysterically high number when expressed as temperature (at the same time that you're also raising the density) Hold it together long enough, don't let it interact with solid matter to cool it, and you've got the holy trinity for fusion. No ongoing heating required.

Also, you don't just easily scale up voltage past several 10s of kV, as you start reaching a lot of material limits for break down (even in vacuum), and engineering gets more difficult for 100+ kV in a small space.

So:
  - Expand the space (which also gives you more plasma volume and thus more power output at a given density), and
  - Keep anything but ionized, under-control, gasses out of the working region

100+ kV is not all THAT difficult to handle in an industrial-sized volume. Air at atmospheric pressure has a breakdown of about 40,000 v/in (though this drops as pressure is lowered). A clean vacuum (except for the working plasma itself) isn't too tough either: Television picture tubes worked fine with no arc-over at acceleration voltages of about a kilovolt per diagonal inch (i.e. 25 kV for a 25" picture tube) and far more than a kV per inch inside the tube. A machine twenty feet across would have substantially lower electric field at 200 kV.

Which is not to say that there won't be issues trying to scale this. But I wouldn't expect anything insurmountable from what you've alluded to here.

This looks familiar from 37 years ago

[InterGuru]

As far as I can tell from the article this looks familiar from 37 years ago.

Check out the Trisops project.

Disclosure: I am the author of the Wikipedia article and a co-author on the cited paper.

Re:This looks familiar from 37 ... CORRECTION

[InterGuru]

CORRECTED LINK
https://en.wikipedia.org/wiki/...

Please captain, help me out to understand...

[Eloking]

Ok, so I've got a huge interest in fusion research. But as a lot of people in /., I don't have enough knowledge to understand how big of a deal it is.

First, 5ms look kinda small, especially when we got no reference as comparison. What was the precedent record? What was the longest fusion was kept "stable" before? Or is it the first time fusion are reached something that could be called a "stable" stabe for "x" time?

Re:Great, but

[rch7]

Only if you build fusion plan right now for the fraction of the cost of fission plant. But it is not going to happen of decades at least. It is highly unlikely that storage will not get many times cheaper by then.

Re:CONgress is screwing up.

[just another AC]

In a breakthrough like this, you "win" by leeching.

As researchers solve the intermediate steps, they will publish (no-one will believe unverified results) in order to get continued funding. This research will not be government top secret so you will only ever be one step behind. One party will make the final breakthrough but others won't be far behind. Yes there could be patents, but that won't stop other governments for something like this.

I'm just glad there are still governments that aren't this cynical (or don't mind paying for world wide benefit).

Re:The Summary, the short version

[I'm New Around Here]

Well, I tried to feed the DC, but had my polarity reversed.