What are disadvantages of nuclear fusion?

[KerbMav]

Maybe more a question for the Whatifs on xkcd, but why not:

Nuclear fusion (as usually proposed) turns hydrogen into helium - while we can make methane from CO2 plus hydrogen and energy to burn it again, nuclear fusion effectively takes hydrogen out of the loop. Granted, it will take time to use up our oceans, but would this not be another finite energy source?

What about the extra oxygen in the atmosphere?

And what influences might the generated helium have?

https://www.iter.org/sci/fusionfuels

"While a 1,000 MW coal-fired power plant requires 2.7 million tons of coal per year, a fusion plant of the kind envisioned for the second half of this century will only require 250 kilos of fuel per year, half of it deuterium, half of it tritium."

So, how much hydrogen per year to generate energy for most of the world? How many liters of water?

Edited May 21, 2014 by KerbMav

[Othuyeg]

Hydrogen's monatomic form is the most abundant chemical substance in the universe, constituting roughly 75% of all baryonic mass.

So yeah, we will have no problem with using up our hydrogen. Also the amount of helium produced would be insignificant, even if we powered the planet with fusion for the next one thousand years. We use alot of helium and it is hard to come by.

http://www.dailymail.co.uk/sciencetech/article-1305386/Earths-helium-reserves-run-25-years.html

[Sirrobert]

Finite doesn't automaticly mean we can spend it all.

Sure hydrogen is finite, but there is so much of it, for practical purposes in this case, it might aswel be infinite

[rtxoff]

Only disadvadvantage i know of is that you need a constant input of energy to keep the fusion running. So if something happens and the fusion power plant has to shut down it will need some initial energy input to get things running again. So you need some backup energy source to restart the reactor. Also the costs running such a plant are probably much higher then other conventional power plants. At the current stage of development they still get no more energy out of the fusion process then they have to input to start it.

[Streetwind]

You don't use basic hydrogen for nuclear fusion. In fact, none of the various proposed reaction schemes use hydrogen as even one of the two partners. Hydrogen is useless for our purposes.

What you use instead is deuterium, and possibly tritium. These things are isotopes of hydrogen (that is, they all have one proton and one electron), but they bring the extra neutrons that are required to form a stable helium core. The makeup of our oceans is thus: 99.985% useless hydrogen, and 0.015% deuterium. Tritium is even rarer, as it's radioactive and decays by itself, and only forms naturally in really tiny amounts in the upper atmosphere. It can be produced synthetically; for example it shows up in fission reactors through radioactive decay of other elements and by transforming deuterium in the coolant water into tritium.

So, downsides?

- You fuse one incredibly rare element with one even more incredibly rare element

- Which is also radioactive

- In a reaction that produces free neutrons, which turns the fusion reactor's components radioactive themselves

There is, however, a more promising option: Helium-3. You can fuse that with deuterium, and get an energy return even greater than that from deuterium+tritium fusion. Also, neither deuterium nor helium-3 is radioactive. And finally, the fusion reaction does not produce spare neutrons, thereby avoiding the irradiation of the reactor.

Downsides?

- Helium-3 is also disgustingly hard to come by because it is rare on Earth, in contrast to other celestial bodies. It is considered so valuable, in fact, that going to the moon for the express purpose of grinding up 150 million tons of solid rock to produce one ton of helium-3 is considered a profitable prospect once fusion becomes a real thing.

But, yeah. The biggest downside to nuclear fusion? We haven't made any net energy return from it yet...

Edited May 21, 2014 by Streetwind

[Red Iron Crown]

Biggest disadvantage is we haven't figured out how to do it efficiently yet.

The fuels are expensive. It may only take 250kg of deuterium and tritium per gigawatt-year, but you have to process an awful lot of hydrogen to harvest those isotopes. Just have a quick look at tritium pricing, you'll see what I mean.

Another problem is it's nuclear. Not a technical problem, but a political one. It would need a more marketable name than "nuclear fusion". "Artificial solar", maybe?

The advantages are great, though. Despite those obstacles, it's definitely worth pursuing.

[phoenix_ca]

When it comes to tritium, currently, it costs $30,000 USD per gram. It's just absurdly rare and difficult to get. A lot of the research with nuclear fusion now is about finding a way to make a reactor that would breed its own tritium supply. That'd simplify the logistical nightmare of feeding a reactor with tritium significantly.

Commercially I'd say the other problem is that nuclear fission will be a very strong competitor for many decades to come. Current generation fission reactors are very safe, reliable, and also have power outputs in the 1,000MW range per reactor. The Advanced Candu Reactor (ACR-1000) does, and that can be fueled with natural uranium, no enrichment required. Sure it needs heavy water to operate, but the cost of acquiring that much heavy water pales in comparison to the cost of acquiring enough pure deuterium to power a fusion reactor. And you get to keep the heavy-water with a CANDU.

Edited May 21, 2014 by phoenix_ca

[mrabinovich]

Tritium is even rarer, as it's radioactive and decays by itself, and only forms naturally in really tiny amounts in the upper atmosphere. It can be produced synthetically; for example it shows up in fission reactors through radioactive decay of other elements and by transforming deuterium in the coolant water into tritium.

The most practical way of producing tririum and fusing it with deiterium is using lithium deuteride for that purpose. It is solid and stable. When a Li-6 or Li-7 atom captures a neutron, it decays into a helium core and tritium core. This reaction is widely used in thermonuclear weapons. Unfortunately, an external neutron source is required, otherwise the reactivity will be below 1, as not every neutron produced by fusion of deuterium and tritium will be captured by a lithium atom. In weapons, the nuclear initiator plays the role of an external neutron source.

[Nuke]

You don't use basic hydrogen for nuclear fusion. In fact, none of the various proposed reaction schemes use hydrogen as even one of the two partners. Hydrogen is useless for our purposes.

What you use instead is deuterium, and possibly tritium. These things are isotopes of hydrogen (that is, they all have one proton and one electron), but they bring the extra neutrons that are required to form a stable helium core. The makeup of our oceans is thus: 99.985% useless hydrogen, and 0.015% deuterium. Tritium is even rarer, as it's radioactive and decays by itself, and only forms naturally in really tiny amounts in the upper atmosphere. It can be produced synthetically; for example it shows up in fission reactors through radioactive decay of other elements and by transforming deuterium in the coolant water into tritium.

So, downsides?

- You fuse one incredibly rare element with one even more incredibly rare element

- Which is also radioactive

- In a reaction that produces free neutrons, which turns the fusion reactor's components radioactive themselves

There is, however, a more promising option: Helium-3. You can fuse that with deuterium, and get an energy return even greater than that from deuterium+tritium fusion. Also, neither deuterium nor helium-3 is radioactive. And finally, the fusion reaction does not produce spare neutrons, thereby avoiding the irradiation of the reactor.

Downsides?

- Helium-3 is also disgustingly hard to come by because it is rare on Earth, in contrast to other celestial bodies. It is considered so valuable, in fact, that going to the moon for the express purpose of grinding up 150 million tons of solid rock to produce one ton of helium-3 is considered a profitable prospect once fusion becomes a real thing.

But, yeah. The biggest downside to nuclear fusion? We haven't made any net energy return from it yet...

im quite fond of the p-b11 reaction, which uses regular run of the mill hydrogen, and regular run of the mill boron, both earthly abundant and with no neutrons.

of course i have a feeling all first gen reactors will be of the d-d or d-t types. this will make the hippies very angry and they will want to ban fusion. we will go through a period like what we are in now with fission where fear mongering results in no power plants being built. instead, continued death by carbon will promptly ensue. until we start to run out in which case we will use the last of our energy on wars. the survivors will be promptly hit by an asteroid 3 decades later, due to being unable to fund a deflection mission.

Edited May 21, 2014 by Nuke

[cantab]

The main disadvantage is that it's really difficult to make work. Hence why decades of research has yet to produce a functioning fusion power plant.

Heck, fusion barely happens in the Sun. The power produced per cubic metre in the Sun's core is about the same as in the tissues of a lizard. The Sun just owes its enormous total power to its enormous size.

[Streetwind]

Huh, I was not aware of that Boron reaction. Interesting! Also, while looking it up, it seems there is also a similar variant using lithium-7 (which forms 92.5% of all naturally occuring lithium) that has an incredible energy yield, twice that of the boron reaction and comparable to deuterium/tritium fusion. Too bad that lithium has the annoying tendency to be spread out everywhere in low concentrations as opposed to concentrated into minable deposits... Still, global production exceeds 35,000 tons per year, so buying a quarter ton lithium per year to run a gigawatt class fusion power plant shouldn't be that much of a supply issue.

Makes me wonder if there are any downsides to using lithium. It's extremely reactive, of course, but that's more of a storage problem (and already solved - store it in mineral oil).

A deuterium/tritium fusion reactor has been calculated to require an energy factor of about 20 (that is, it produces 20 times the amount of energy that is put into it) in order to produce electricity at prices comparable to existing methods. Our best result so far has been 0.7 - quite a steep hill left to climb yet! However, it's mostly the ridiculous cost of tritium that drives the need for a high energy factor. If you could use much cheaper lithium and hydrogen, maybe you could be commercially viable at energy factors as low as 2-3?

Edited May 21, 2014 by Streetwind

[Kryten]

im quite fond of the p-b11 reaction, which uses regular run of the mill hydrogen, and regular run of the mill boron, both earthly abundant and with no neutrons.

Even if you manage to produce pure ¹¹B as fuel, you're going to get neutrons from the ¹¹B + p → ¹¹C + n and ¹¹B + α → ¹⁴N + n side reactions. Then there's the issues with power density and Bremmsstrahlung losses.

[Seret]

It will still generate a fair amount of low- and mid-level nuclear waste, due to the high neutron flux in the reactor. It won't generate the more troublesome high-level wastes you get in fission reactors, but it'll still need to be disposed of in specialised waste streams.

A containment failure could be pretty frisky too, you'd want to make damn sure there was no chance of venting your nasty plasma to atmosphere.

[Streetwind]

Fusion containment failures are less of an issue than you might think. That plasma has abysmally low densities at atmospheric pressure; it's only reactive because it is kept under gigantic pressures. Even if all of a reactor's plasma was to escape into the atmosphere at once, it would be less than a gram of material all in all, only a tiny fraction of which would be radioactive.

Then there's the fact that the fusion chamber is a vacuum. Plasma will not escape into the atmosphere - much rather, the atmosphere will enter the breached chamber and snuff out the fusion reaction in an instant, much like a candle in a hurricane.

[Nuke]

even if you loose magnetic confinement or whatever, you still have the vacuum chamber. loosing confinement still results in negative pressure with respect to the outside environment.

[m4v]

Well, currently its disadvantage is that we can't build one, there are great unsolved technological difficulties, and if these are solved likely the cost will be its main disadvantage.

Still this doesn't stop people from handwaving all that and say that we will have cost effective fusion reactors in just 50 years.

[Seret]

Surely if you lost magnetic confinement the plasma would impinge on the walls of the vacuum vessel, which also presumably contains the coolant? That seems like it could be pretty explody. I presume they're designed with a containment building around them for just this reason, similar to a fission reactor. ITER's website simply mentions a "reinforced" tokamak building, but it's unclear whether that's to stop stuff getting in or out.

[Streetwind]

ITER will be running deuterium/tritium fuel so they need heav shielding to block the neutron radiation.

But really, fusion reactors cannot explode. It's physically impossible. A nuclear fission reaction is a self-perpetuating, runaway chain reaction that will proceed to the point of explosion if left uncontrolled; a nuclear fusion reaction, on the other hand, is an incredibly fragile thing that you have to force, with the utmost of effort involved, to occur. It takes an object the size of a star in order to create a self-perpetuating fusion reaction. Inside our tiny reactors, any deviation from nominal operation parameters simply results in the reaction dying instantly.

- If the magnetic containment failed and plasma touched the walls of the vacuum chamber, it would heat that wall by a couple degrees at best while losing all of its energy. This is, again, due to the fact that the plasma has virtually no density without being confined under extreme pressure. It doesn't matter if it's a million degrees hot; if there's only one gram of mass, then that is enough energy to heat one million grams (one ton) worth of vacuum chamber wall by exactly one degree. The mere act of touching that vacuum chamber wall will drain so much energy from the plasma that the fusion reaction dies instantly, if it isn't already dead from the loss of pressure before it even reaches the wall.

- If you injected a thousand times the intended amount of fuel, the magnetic confinement would simply fail to be able to contain it, and no fusion would happen in the majority of the fuel (and the contamination would likely snuff out the fusion reaction too).

- If the vacuum chamber is breached, air rushes in, disrupts the reaction and instantly saps all heat from the ephemereal plasma, again snuffing everything out instantly.

Now, a fusion reactor can implode, because there is a vacuum chamber involved, and implosions can cause quite significant shockwaves too. Technically a major implosion event could wreck a reactor building and (if it is running a neutron-emitting fuel mixture) toss irradiated reactor parts across the surrounding hundred meters. But that scenario is so unlikely, we have a bigger chance to be saved from a planet-killer asteroid impact by a rag-tag team of oil drillers with nuclear warheads. You can probably engineer vacuum chambers with fault lines so that implosion events don't take the entire chamber at once, too.

[Ralathon]

You don't use basic hydrogen for nuclear fusion. In fact, none of the various proposed reaction schemes use hydrogen as even one of the two partners. Hydrogen is useless for our purposes.

What you use instead is deuterium, and possibly tritium. These things are isotopes of hydrogen (that is, they all have one proton and one electron), but they bring the extra neutrons that are required to form a stable helium core. The makeup of our oceans is thus: 99.985% useless hydrogen, and 0.015% deuterium. Tritium is even rarer, as it's radioactive and decays by itself, and only forms naturally in really tiny amounts in the upper atmosphere. It can be produced synthetically; for example it shows up in fission reactors through radioactive decay of other elements and by transforming deuterium in the coolant water into tritium.

So, downsides?

- You fuse one incredibly rare element with one even more incredibly rare element

- Which is also radioactive

- In a reaction that produces free neutrons, which turns the fusion reactor's components radioactive themselves

There is, however, a more promising option: Helium-3. You can fuse that with deuterium, and get an energy return even greater than that from deuterium+tritium fusion. Also, neither deuterium nor helium-3 is radioactive. And finally, the fusion reaction does not produce spare neutrons, thereby avoiding the irradiation of the reactor.

Downsides?

- Helium-3 is also disgustingly hard to come by because it is rare on Earth, in contrast to other celestial bodies. It is considered so valuable, in fact, that going to the moon for the express purpose of grinding up 150 million tons of solid rock to produce one ton of helium-3 is considered a profitable prospect once fusion becomes a real thing.

But, yeah. The biggest downside to nuclear fusion? We haven't made any net energy return from it yet...

You underestimate how much deuterium there is on the planet. One in every 6400 hydrogen atoms is a deuterium isotope. This means that if you filter an olympic swimming pool for heavy water you can extract a large bathtub full of the stuff. That's enough to keep a nuclear fusion reactor running for months. The oceans contain enough deuterium to keep us running for millions of years, even if we take into account the exponential power increases of the past centuries. Furthermore, unlike uranium it is very easy to separate deuterium from hydrogen thanks to the large weight difference.

In fact, it is so easy you can do it yourself with some basic equipment: Electrolyse some water and let the hydrogen diffuse through a long pipe. Since hydrogen is twice as light as deuterium it'll escape way earlier. Of course the yield on this is going to be pretty crappy, but it demonstrates the principle.

He3 farming on the moon is not worth it by the way, if you want to farm He3 you should go to Saturn. But I recommend you just make it at home, just bombard some tritium with neutrons and you'll get He3. This is how practically all He3 is made right now.

[Elthy]

The dangers of fusion are not comparable to fission...

First, a core-meltdown is impossible. There is only about 1g of plasma in you reactor, when it gets in contact with the hull it will cool instantly, so it cant melt through the hull since it doesnt have enough heat energy. Also the is not heatproduction after a shutdown, so you dont need complicated backupsystems for blackouts.

Second, the radioactive waste is not (that) dangerous. While fission waste will be very radioavtive for millions of years the fusion waste will be dangerous for about 100 years, which is controllable.

[Idobox]

There is a much cheaper way to get He3 than mining the few ppm you can find in moon rocks: let tritium decay.

Also D-He3 fusion might be aneutronic, but you will get a significant amount D-D reaction on the side, and these will release plenty of neutrons, plus tritium nuclei that will fuse and release more neutrons. Its main interest is the higher yield.

D-He3, D-Li6 and p-B fusion will not have commercial applications for a long time because they are even more difficult to achieve than D-T.

And to produce tritium, you need to expose lithium to neutrons. With lithium 6 you get one tritium and 4.8Mev, with lithium 7, you get one tritium and another neutron. Since D-T fusion generates plenty of neutrons, surrounding the chamber with lithium 7 could lead to several tritium nuclei produced for each consumed one.

You can also keep some He3 around: it absorbs neutrons easily to turn back into tritium.

[Aghanim]

Here is the thing: (Un)fortunately, fusion reactor could not explode because we couldn't even reached ignition in the first place! Assuming we are using magnetic confinement, like tokamak or magnetic mirror

Inertial confinement fusion uses explosion, but it's very small

Now, hydrogen-boron fusion is kinda neutron-less, but there's still deadly gamma radiation.

SPAM warning!

Actually, I prefer dense plasma focus more than current magnetic confinement setup, as it could use p-B11 fuel and making a propulsion system out of it is pretty much straightforward. However, the best thing is huge megajoule DPF and small kilojoule DPF have the same plasma density, but of course the larger ones have more power. This is interesting because this allows us to miniaturize fusion reactor to as small as we want, until the capacitor and vacuum pump losses outweigh the power generated

Link to its website: http://www.focusfusion.org/

[Winter Man]

p-B11 is all very well and good, but it requires a much higher temperature to achieve. Several orders of magnitude higher than D-T, that is. I'm all for it, but the first to market will likely be the method of the future, which is looking to be General Fusion's steampunk piston driven machine (running on D-T). At least for a generation or two and only in large scale applications, that is.

[Sirrobert]

There is a much cheaper way to get He3 than mining the few ppm you can find in moon rocks: let tritium decay.

Also D-He3 fusion might be aneutronic, but you will get a significant amount D-D reaction on the side, and these will release plenty of neutrons, plus tritium nuclei that will fuse and release more neutrons. Its main interest is the higher yield.

How does that work?

Tritium is 1 proton and 2 neutrons. He3 is 2 protons and 1 neutron.

How the hell can tritium decay into He3? Or am I missing a step?

[Kryten]

How does that work?

Tritium is 1 proton and 2 neutrons. He3 is 2 protons and 1 neutron.

How the hell can tritium decay into He3? Or am I missing a step?

A neutron converts to a proton through antielectron/positron emission.