China's Fusion Reactor Reaches 100 Million Degrees Celsius (abc.net.au)

hackingbear shares a report from the Australian Broadcasting Corporation: The team of scientists from China's Institute of Plasma Physics announced this week that plasma in their Experimental Advanced Superconducting Tokamak (EAST) -- dubbed the 'artificial sun' -- reached a whopping 100 million degrees Celsius which is six times hotter than the core of the Sun. This temperature is the minimum required to maintain a fusion reaction that produces more power than it takes to run. The Chinese research team said they were able to achieve the record temperature through the use of various new techniques in heating and controlling the plasma, but could only maintain the state for around 10 seconds. The latest breakthrough provided experimental evidence that reaching the 100 million degrees Celsius mark is possible, according to China's Institute of Plasma Physics. "While the U.S. is putting new restrictions on nuclear technology exports to China, inventions and findings of EAST will be important contributions to the development of the International Thermonuclear Experimental Reactor (ITER)," writes Slashdot reader hackingbear. The reactor is currently being built in southern France with collaboration from 35 nations. According to the Australian Broadcasting Corporation, it is expected to be "the first device to consistently produce net energy, producing 500 megawatts of clean and sustainable power."

Comments

Still useless for energy production

[Antique Geekmeister]

I'm afraid that all deuteriam and tritium based fusion reactors rely on fuel that is in extremely limited supply, especially tritium. Since the main source of tritium on Earth is nuclear decay from fission reactors, if there are enough fission reactors to generate enough of the very inefficiently used fusion fuel to generate significant, they can generate many times more energy from the fission reactors without having to engage in dangerous refinement of the tritium.

It's theoretically possible that thallium, which is much more plentiful than hydrogen isotopes, can be used for susion. But I'm sad to say that hydrogen fusion _cannot_ be effectively used for energy. Every technology that harvest or generate enough of the hydrogen isotopes manages and can harvest so much other energy that hydrogen isotopes re only a useful research byproduct, not a comparable energy source.

Re:Sun's core too cold for fusion, sort of

[timeOday]

Oh, that reminds me when I asked my chemistry teacher why water would evaporate even below the boiling point. He said something similar, the temperature is the average but on occasion a molecule gets enough energy to exceed the threshold (thus cooling the others when it leaves with its heat). Similar? Or not?

But how much is that in electron volts?

[Ungrounded Lightning]

China's Fusion Reactor Reaches 100 Million Degrees Celsius

Plasma energy sounds really large when you express it in temperature. But a more convenient gauge may be the voltage needed to accelerate the particles to velocity magnitudes correspondng to that sort of energy. This is also directly applicable to fusion systems, such as the Farnsworth-Hirsch or Bussard's Polywell, which use electric fields to accelerate the particles into the reaction volume.

Both electrons and hydrogen nuclei have a charge magnitude of 1, so dropping them across a potential difference of N volts adds N electron volts of energy to each particle. Then, if you let the plasma thermalize to a Maxwellâ"Boltzmann distribution, the electron temperature will be (by definition) the temperature of the distribution is about 2/3 that corresponding to the average electron energy.

So to go from degrees Celsius degrees (of a thermalized plasma) to electron volts:
  - Subtract 273.15 - a .003% drop in the bucket. (Kelvin step sizes are the same but Celsius starts at 273.15 Kelvin.)
  - Divide by 11,605 to get electron volts.
  - Multiply by 2/3 to get the average energy of the electrons and ions.

That's an acceleration voltage of 6,025 volts (or 9,037 if you're going to react them before they thermalize). That's right in the ballpark for high-end vacuum tube technology - like the second anode on a CRT. (Those ran about 3000 to 6000 V in the 1940s, and about 25,000 V when modern color tubes were being replaced by flat panels.)

You can see why we all had high hopes for things like Polywell, where (if it worked as expected) a "gassy vacuum tube" that would fit in a strip-mall store's back room, with all supporting equipment (mostly mid-20th-century style electronics), and provide 100 MW of DC at cross-country power line voltages.

Of course many of the other methods for directly heating plasma heat the electrons much more than the ions. So the average energy of the plasma may be substantially lower.

Re:Sun's core too cold for fusion, sort of

[Anonymous Coward]

Similar in that statistically unlikely things happen quite often with enough time or space.
The mean free path of a neutrino is calculated to be several light years through solid lead before hitting a particle.
However neutrinos are emitted by the sun so frequently and neutrino detectors are so large that we can detect them reasonably frequently.

ITER wont produce power

[angel'o'sphere]

It will run at 400 - 600 seconds and will produce more energy than it consumes, that is all. There is no power plant attached nor will there ever be: https://www.iter.org/sci/Goals

And the power production is not clean as long as we use deuterium + tritium, the reactor vessel will have to be replaced around every 10 years and discarded as highly radioactive waste.

Regarding sustainability: ITER will attempt to breed tritium ... lets see how good that works. Otherwise we had to farm tritium from the sea, which is energy intensive and causes another spot in the chain to work with an radioactive element.

Re:Gravitational plasma confinement/optical densit

[dryeo]

My understanding is that the energy output, per cubic meter, is about the same as the human body, 50-100 watts or whatever. Just that there are a lot of cubic meters in the core of the Sun, so it adds up. As the AC says, proton-proton fusion is slow, even at the pressures and temperatures at the core.

Re: could only maintain the state for 10 seconds

[jd]

Not really. The only direct products you make will be Helium-4 (stable), Helium-5 and Helium-6. You could smash up or change isotope a carbon, nitrogen or oxygen atom, I suppose. But you're talking very short half-lives.

The concrete is a problem. Fortunately, the Iranians have a recipe that is less likely to powder or fail. So, with trade restored under the joint agreement, we're ok.

Oh.

Re:Sun's core too cold for fusion, sort of

[Anonymous Coward]

The thermal energy produced per cubic meter in the core of the sun is comparable to a compost pile and less than per volume heat produced by a human. The Sun is just really, really big, so emitted light gets re-absorbed as heat, and even a relatively conductive material makes a decent insulator if thick enough. The slow fusion process of the Sun can get as hot as it does just because the heat is so well trapped.

On Earth, we are limited to only a couple meters of insulation, instead of 100,000s of km. The reactors will lose heat many orders of magnitude faster than the Sun, so they need to produce heat much faster. Luckily DT fusion is much faster than pp fusion, and the reaction rate scales up quickly with temperature too. So with a temperature 10 times that at the center of the Sun, with a better fuel choice, you end up with a much faster reaction that can still keep hot despite the much less insulation.

Also, because you need to make heat faster than it leaves, usually just the temperature is not enough. The triple product is a common metric, where you multiply the temperature, density, and confinement time (how long a typical particle or parcel of energy sticks around, not the lifetime of the plasma) together. You need it to be hot enough to fuse, you need enough fuel at that temperature to get enough reactions, and you need it to stick around long enough before carrying heat away. There is some room for trade off between the three. This metric has been scaling up over the years in a pattern similar to Moore's law, because of improvements to confinement time and density (temperatures haven't changed much at this point).

That can partially explain how you can have fusion temperatures, but not self-sustaining from reactions, as their confinement time and density might be on the low side. Also, a lot of experiments run with DD instead of DT, as it behaves essentially the same, but you don't have to deal with as many neutrons and you don't have to deal with handling radioactive tritium. JT-60 has already made DD plasmas that would produce more power out than in (for a short time) if they had been DT instead, and there is no doubts about the DD plasma being any different than DT in that case. (There will be some difference at higher reaction rates, as about 80% of the energy of a DT reaction leaves the plasma as a neutron, so when the fusion power is about 5x what is being put into the plasma, that other 20% trapped power will be comparable to external heating, and there would then be an advantage to using DT fuel.)

Re:Apparently not

[mikael]

They keep running into problems. I've read a few papers, and they would hit problems such as the metals used weren't strong enough to withstand the magnetic fields they were generating. That was fixed. Then the plasma rings would start to twist, buckle, warp and pinch into singularities. Stellerators fixed that problem by putting some torsion into the plasma rings. Tokamaks fixed that problem by adding extra magnetic field randomness or something to break up the standing waves. That fixed that problem. Then the neutron bombardment started poking holes in the metal structure, which weakens it over time. Maybe that has been fixed, but it keeps going round and round.