Extracting power from a fusion reactor

I watch a lot of documentaries on various subjects, and one thing I’ve heard a couple of times recently and had stuck in my brain is how much more power could be generated from a fusion reactor as opposed to the current fission reactors.

How would that be reflected in terms of actually using that power? As I understand it a fusion reactor generates power the same way as fission in using the reaction to heat water to spin a turbine right? Or is there an alternative method to better harness a fusion reaction?

I don’t know if it’s better from a practical point of view, but there have been experiments in converting the energy directly into a voltage without going through a fluid heat exchanger and turbine setup.

I’m guessing that the reason they were saying more power could be generated from fusion is more to do with what happens when you scale things up and have a lot of power plants running rather than talking about specific reactor designs. With the current fission power, no one wants a new plant in their back yard, and the fuel supply is limited (there’s only so much uranium extracted and refined each year). A fusion plant would basically run on water (not exactly a scarce resource here on earth) and would not have the nuclear waste issues, meaning that there’s no competition for scarce resources when you scale things up.

Not just experimental
These guys are in actual use…
wikipedia entry for Radioisotope thermoelectric generator

The Radioisotope thermoelectric generator is a no moving parts power supply based on a source of heat… for long life with no maintenance required, they have used fission.

Terribly inefficient of course, 5% heat to power… but they power satellites… and remote equipment.

So if you had a long life zero maintenance fusion process, it could be the heat source for a thermocouple power supply.
And building heater too … where it would excel … (could be up to 100% efficient .eg to heat a building at the poles… )

My understanding is that most of the energy released in the deuterium-tritium reaction is in the form of high-energy neutrons, and that the only practical way to extract this energy is to let the neutrons heat a thick layer of absorbing material. So we’re going to use a hundred-million degree plasma to make steam. :rolleyes:

In fact a pure fusion reactor may not be commercially viable. One proposal is a hybrid fusion-fission reactor that uses the high-energy neutrons to fission (normally unfissionable) Uranium-238, the cheap common isotope. Unlike a conventional fission reactor this couldn’t meltdown because only the fusion neutrons keep it going, and radioactive waste could mostly be converted into short-lived isotopes.

Yes, the problem with fusion is the output is mainly neutrons. (Incidentally, this is the problem with “cold fusion”. If fusion s actually happening to produce the heat they claim, there would be a dangerously high level of neutron emissions. SO far nobody has reliably detected this in a reproducible experiment).

The old story about “Fusion is cleaner because there’s no waste problem like with uranium fission”- the blanket that stops the neutrons would quickly become very radioactive. One suggestion was to use the neutron bombardment to enrich non-radioactive U235 to U238, thereby creating more active fuel for fission reactors.

This is not direct conversion, it is the standard conversion to heat first then to another form. And that is the problem once you convert to heat you lose a lot of usable energy.

Direct conversion would convert the reaction energy directly to voltage.

This is the fundamental problem with fusion. It probably can’t ever be economically viable for power generation.

Suppose we get a working research reactor that produces net positive power in the reaction itself. What’s going to be in this baby? It’s going to have, at a minimum, thousands of miles of superconducting cables, building sized banks of lasers, and so on. One horrible problem here is that superconducting magnets need to be kept cool (probably using liquid helium), and yet somehow you have those magnets next to a neutron source that is heating everything back up, creating steam.

Similarly, lasers are not very efficient, so even if you have positive gain in the fusion reaction, you’re driving a laser that only converts a few percent of input energy to light with energy from a heat engine that is also limited in efficiency.

So you’d need huge fusion gains for it to even work at all. Ok, so let’s posit they find a way to get these gains.

How is it going to ever be economical to build a power station that requires immense quantities of high grade components, immense quantities of rare elements like rare earths, and the whole thing has to be aligned and tested by crews of PhD physicists.

Contrast that to something a little simpler. We could just make a factory that creates solar panels completely autonomously, reducing labor costs to near zero. We’d build a second factory that does the same thing for lithium iron batteries. Solar panels and lithium iron batteries are far simpler devices than anything that goes in a fusion reactor, and they don’t require more than minimal training or education to use.

In the far future, you’d just make solar cells directly and put them in space so you don’t need batteries at all. (you’d make the solar cells in a robot factory on the moon and launch them with a mass driver for minimal cost). This effectively is just building the energy collection part of a fusion reactor, and letting the sun do all the rest of the steps. It’s always going to be cheaper than building a fusion power reactor, no matter how advanced future technology becomes.

With all that said, there is a form of fusion called “aneutronic fusion”. It requires certain extremely difficult fusion reactions. However, you get beta particles (electrons) directly as the output. You slow the moving electrons down between 2 charged meshes and get power out directly.

Also, there’s at least 1 use case for fusion : rocket engines. You don’t try to generate electricity, you just let the high velocity particles from the fusion reaction fly away from your spacecraft. This gives you better thrust/weight than generating energy and then powering electric engines (such as ion engines or VASIMIR). It’s less fuel efficient, though. So the only use is manned missions inside our solar system (the semi-bad fuel efficiency means you’d never get enough velocity for interstellar missions)

See here : http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20050160960.pdf

So when we finally get that fusion reactor running everybody will have a big celebration and then stand around it saying “Well, now what do we do?”

Pretty much. I mean, let’s just assume we have the end state of manufacturing technology. We can just click a mouse and molecular printers in a vacuum chamber will make anything. Even then, I suspect fusion wouldn’t be economically viable, because the raw materials to make a fusion plant would exceed the raw material cost for other sources of power generation. (like solar or fission)

One angle worked on is a Tokamak design - a giant donut magnetic field that squeezes super-hot plasma. Again, miles of superconducting wire feeding giant magnets, but probably more efficient than lasers. Still have the problem that the liner is what will absorb the neutrons and produce heat, becoming radioactive in the process.

I don’t know if fusion can be made economically viable - that’s why governments are spending money on it - but your alternative is at least as uncertain.

You say fusion is not viable as it would use “immense quantities of high grade components, immense quantities of rare elements like rare earths” but solar panels require rare materials - normally tellurium - to manufacture. Each panel only uses a fraction but if they are going to make a serious impact on the world energy needs you need a LOT of panels. And that’s another problem - taking Britain which is at high latitude and is frequently cloudy, you would need to cover 5% of the total land area of the country - cities, farms, hillsides - with photo-voltaics to make a dent in our energy needs. Also, you say a fusion power plant would need to be set up and tested by PhD physicists, actually no, a production power plant would need good engineers to set it up and technicians to run it. the physicists are needed for the research not the operation just as with a “a factory that creates solar panels completely autonomously”.

Finally, building and operating robot factories on the moon and beaming power back from space is way more out there than developing fusion as a power source - the costs and risks are literally astronomical.

Fusion between hydrogen and Boron-11 releases three high-energy alpha particles (helium nuclei) moving at high speed as the only product of the reaction. You can then decelerate these through a series of electrical meshes, resulting in direct electrical output of about 14 million volts DC from your reactor. That can then be turned into usable electrical power with off-the-shelf high voltage electronics.

Unfortunately, hydrogen - Boron-11 fusion is incredibly difficult, and may be impossible to achieve at all in a way that yields more energy than it takes. But if we do figure out a way, we’ll be able to do the output energy conversion without using steam turbines.

It’s interesting to note that if the 2.4 trillion dollars that the US will have spent on the Iraq and Afghanistan wars had been spent on Solar energy installations, it would have provided 800 GW of electrical capacity, or enough to power 80% of the US!

(Assumptions - Solar energy cost of $3/watt installed (overestimate), Total US peak electrical generating capacity of 1,000 GW).

just to sidetrack even further, the problem is that solar power, and wind and other energy sources, is really waiting for a good storage medium, a reliable deep-discharge, multiple-cycle battery or large-capacity capacitator. Like fusion power, this is something that is always a few years away.

(But we’re getting there. Modern electric cars, for example, benefit from far better batteries than 20 years ago.)

Well while they are standing around scratching their heads on “how to do fusion”

they also thought "Well , what sort of fusion would make the lowest maintenance reactor ? I mean, if we make neutron’s , we’d pretty quickly transmute the shielding into a pile of poisonous dust ! So we’d really want to be trying to reduce the number of neutrons created… "

Therefore… the topic of Aneutronic fusion… fusion that doesn’t make neutrons.
Wikipedia on Aneutronic fusion
Note that the fusion of p+B would also make 4, 12, and 16 MeV gamma rays… You’d be able to detect the operation of the reactor from the moon…