The Bizarre Reactor Scientists Hope Will Save Fusion Research (sciencemag.org)

sciencehabit writes: In a gleaming research lab in Germany's northeastern corner, researchers are preparing to switch on a fusion device called a stellarator, the largest ever built. The €1-billion machine, known as Wendelstein 7-X looks a bit like Han Solo's Millennium Falcon, towed in for repairs after a run-in with the Imperial fleet. Stellarators have long been dark horses in fusion energy research but the Dali-esque devices have many attributes that could make them much better prospects for a commercial fusion power plant than the more popular tokamaks: Once started, stellarators naturally purr along in a steady state and they are not prone to the potentially metal-bending magnetic disruptions that plague tokamaks. Unfortunately they are devilishly hard to build.

Comments

Title is misleading

[benjfowler]

There's nothing "new" about the stellarator at all.

I'm pretty sure that Lyman Spitzer came up with the idea at Princeton before the Russians did at the Kurchatov Institute. The only reason why the tokamak is more famous, is that the physics performance (particle, energy confinement) was for the longest time, way better in tokamaks (and may well still be). Also, tokamaks are way easier to build (but harder to operate).

That said, I've read suggestions that stellarators might be able to be optimised in ways that are impossible in tokamaks, pending further breakthroughs. The machines will still cost a fortune to build though -- and cost is going to be a BIG barrier to adoption of fusion as a power source at any rate.

Re:Title is misleading

[knightghost]

We blew $5 trillion on useless wars and $20 trillion so that wall street could get some short term bonuses. A billion dollar fusion engine isn't even a rounding error.

DOE report says fusion is likely uneconomical

[PeterM from Berkeley]

Went and looked for answers to my own question:

This report from DOE
http://web.ornl.gov/~webworks/...

has figures showing that they forecast the cost of fusion power to be between 68 to 80 "mill/kWh", (apparently mills are thousandth's of a 1999 dollar) which is more expensive than any alternative they examined. Wind power they forecast to cost between 20 to 40 "mill/kWh".

If the people at DOE who wrote that report are good forecasters, then fusion is DOA. Alternatives will be less expensive.

Yes, you can make "technology advancement" arguments that the DOE forecasters are wrong, but the cost of wind and solar generators are dropping all the time, too, and storage options might get radically cheaper as well. I think investment in solar + wind + storage actually dwarfs investments in fusion, so the market seems intent on fulfilling DOE's prophesy.

Fusion may really only come into its own when we go live in the asteroid belt or the outer solar system.

--PeterM

Re:DOE report says fusion is likely uneconomical

[CaptainLard]

Whether a power source is economically viable is by no means static.

Can't argue that but this brings up an interesting question (for me anyway). What could make wind and solar more expensive in the future? The raw material costs are basically nothing compared to fossil fuels (where they are essentially basically 100%). There is plenty of silicon just lying around on the surface. Some of it may be easier to process but I think that problem is largely solved.

I don't know what the global lithium supply looks like but I do know LiIon batteries can be recycled. The reason China has all the rare earth mines is because their government subsidized them to corner the market in the short term. Rare earths aren't really that rare.

For wind you need resins and what not but again, the material input is minuscule compared to the lifetime energy output. You mentioned real estate and places that make sense....but there are a LOT of those places where it does make sense. Often the real estate is pretty cheap (deserts, etc) and hey, the solar panels covering just a little over 1/3 of my roof provide 110% of my yearly electricity use! (yeah storage...that's being resolved faster than even I thought it would)

Given all that I know about wind and solar it seems that if anything, prices will only get cheaper.

Re:Even if ITER or W7X works, is it economical?

[david_bonn]

For all of this, in the very best case W7X will only sustain fusion for thirty minutes (according to Wikipedia). That is an extremely long way from being practical.

Even assuming it works very well, we are an extremely long way from solving all of the problems required to build a practical working fusion reactor.

Some of the problems remaining to be solved:

• Neutron flux (part 1). Most of the energy from the deuterium-tritium reaction is in the high-energy neutron produced by the reaction. The best estimate is that the neutron flux from a 1GW fusion reactor would be one or two orders of magnitude higher than from a fission reactor. No known material can withstand that neutron flux. One other way to look at it is that in five years of operation, every atomic nucleus in whatever radiation shield you build will be hit hundreds of times over a five year period.
• Neutron flux (part 2). the deuterium-tritium reaction produces one neutron. That neutron has to (1) heat a working fluid that can be used to run a turbine, and (2) strike a lithium nucleus with enough energy to breed tritium. You need to do that with every damned neutron to have a self-sustaining system. This is made even more challenging by the fact that neutrons will be emitted isotropically from the reactor. Yes, there are materials that can act as neutron amplifiers, but no one has ever done that on a large scale and it probably won't be easy or simple.
• Lithium. You are going to need a lot of it. A 1GW reactor will probably need around 10000 tons of lithium. At $7/kg, that is seventy billion dollars worth of lithium. That is also a significant percentage of the world's annual production of lithium.
• Tritium. Once you've made the tritium from the lithium, you need to get it back into the plasma where it can do some good. I note that both tritium and lithium will easily react with each other and separating them will be tricky.
• Helium removal. Your fusion reaction will produce helium. Too much helium in the plasma will interfere with the reaction and lower the efficiency of the reactor. You need a system to get the helium out of the plasma without cooling it down. This system must operate continuously.
• Scaling. W7X has a plasma volume of around 30 cubic meters. A 1GW fusion plant would need a plasma volume on the order of 1000 cubic meters. W7X will cost around a billion dollars -- straight-up extrapolation implies a cost north of 30 billion dollars. That doesn't include all of the systems described above or a turbine to actually generate electricity. I also point out that scaling up isn't necessarily cheaper either.

I'd also note that solving each of the above problems is not going to be cheap. It is hard to imagine how a fusion plant can be made for less money than an existing fission plant, and those plants are already not competitive. Chances are it would be better and cheaper to build lots of batteries with all that lithium and a lot of wind turbines and solar panels. That would get you the same amount of energy, probably.

Sources: matter2energy, Do The Math