Questions about fusion safety, waste

If polywell fusion is developed, in what ways will the world change for better or worse? Discuss.

Moderators: tonybarry, MSimon

Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Questions about fusion safety, waste

Post by KitemanSA »

This topic replicated from "News".
jsbiff wrote:Hello all.

In light of the recent events in Japan, I wanted to double check my understanding of fusion power safety characteristics.

This has been my understanding of fusion vs. fission, thus far:

In a fission reactor, you typically have a large fuel supply deposited into the reactor, once every 1-5 years (except for naval reactors which use higher-enriched uranium, and can go something like 10 years on a fuel load). This amounts to tons of material - most of which is not radioactive, but about 1-3 percent at any given time will be radioactive, and very highly radioactive immediately after reactor shutdown. Decay heat continues to generate a small fraction of the power of the reactor during regular operation, but since the reactor generates 2-5 Gigawatts of thermal energy during normal operation, that 'small fraction' is still over a Megawatt for a few weeks or month.

Additionally, the nature of the fuel is that it has a natural tendency towards fissioning on it's own - it doesn't need any active, outside forces, it simply needs enough fuel, in the right physical configuration, with moderator (like water) between it, then it will just spontaneously start fissioning (I believe to "start" a reactor they just withdraw the control rods, and the process will self-start?). Finally, the fuel has radioactive fission products which will continue to be significantly radioactive for several years, followed by weakly radioactive for centuries (a so called "long tail", e.g. asymptotically approaching zero).

Now, a fusion reactor, on the other, will generally be 'fed' fuel on a just-in-time basis, much like an internal combustion engine, which 'sips' gas or diesel as it operates. So, the amount of fuel in the reactor at any time is tiny. Fusion fuel doesn't even want to fusion - to achieve the fusion, you have to actively apply external forces to the nuclei to bring them together so violently that they have a good chance of fusioning.

The general approach to do this is by super heating the fuel into a plasma, either using a magnetic field (e.g. tokamak), or by using magnetic fields to trap a lot of electrons into a virtual (cathode? or is it Anode? Never can keep them straight), that is, a small, highly negatively charged electric field in space, then inject the fuel nucleii into the reactor (polywell), such that the strong negative field powefully attracts the nucleii towards the center of the node, super-heating the nucleii into a hot plasma, where the fusion can, hopefully, happen.

Questions: Are any of the fusion products of the proposed fuel mixes, radioactive. If they are radioactive, are they strong emitters, or weak emitters (e.g. Tritium is a weak emitter). What kind of half-lives do they have? I believe that, essentially, even if a fusion reactor exploded, there'd be very close to zero radiation released into the environment, in every possible case, because the amount of radioactive materials would be very small?

I believe I read somewhere that neutrons from the fusion reaction could be absorbed by some other materials, and make those other materials radioactive, but I think such 'activated' materials would be a fairly weak source of radioactivity?

If a disaster hits a polywell reactor, as an example, what would likely be the consequences? I believe that because fusion requires an active, external force to be applied, you can instantly (e.g. within microseconds or milliseconds) shut off the magnetic fields and the plasma will immediately dissipate? My understanding is, the plasma, during operation, is millions of degrees - does that pose any risk? My understanding is that there is such a tiny amount of plasma in the reactor at a time, that as it begins to interact with nearby matter (any air which might be in the reactor, reactor walls, etc), it will extremely quickly (within seconds) cool down to 'ambient' temperature, perhaps leaving the reactor at a temperature of a couple hundred degrees or less (because the mass of the reactor is hundreds of thousands or millions of times greater than the mass of the plasma)?

I believe once the fusion shuts down, there's no significant amount of decay heat, so that the reactor will cool off within like an hour or two, to 'room temperature'?

In a reactor which uses hydrogen fuel, there might be a small risk of hydrogen explosions if fuel lines/tanks got damages, and hydrogen started leaking, but it would be plain, old, 'boring' combustion which would only be a threat to the buildings and people on the plant premises?
To which
pfrit wrote:First, this should be in general, not news.
Second, Neutron activation is no joke. It can make some really nasty stuff. Granted that this happens in a fission reactor as well, though my understanding is that it not as severe in fission as the neutron flux is not as intense. Also since the neutron flux would be higher, the need to build everything out of neutron safe materials would greater.
Assuming we are talking about a system that uses SC magnets, when one of these thing pop, they will pop awfully hard. Probably without breaching containment, but it would explosively destroy itself. I would expect no damage outside the reator building, but it would be truely trashed and this would be expected to be much more likely than a fission incident. However since the flux would be higher, the operators would be very well protected from any accident as the main neutron sheilding method (distance) would protect them from any explosion.
So, to sum up, much more likely to fail and expensive failure modes, but very low chance of harming anyone and move this to general.
Where upon
shipjack wrote:I think it depends on the type of fusion reactor.
There will be much fewer fast neutrons in a PB11 burning reactor, so also less heavily irradiated reactor parts.
A reactor like the Dens Plasma Focus (if it works) would also not have any magnets that can explode.
I am wondering how much better or worse a Fusion- Fission- Hybrid like the guys at Helion are proposing, would have worked.
This thing is actually burning nuclear waste...

Oh and pfrit, compared to all the political crap that we have had in "news", I think that this is tollerable.
and in reply
jsbiff wrote:I don't think I can move the post. I thought about putting this in general, but since it's prompted by the news from Fukushima, I then thought it might be better in here. Basically, I wasn't really sure where to put this, and news seemed like as good a place as any.

Back to the question at hand, I just had a thought - with 'normal' electromagnets, you shut off power to them and their magnetic fields decay extremely quickly (to human perception, instantly).

Super Conductors, I believe, get a loop going, and because there is no resistance, can basically act a lot like 'permanent' magnets. I guess the 'shutoff mechanism' in that case is just to stop injecting fuel into the reactor, and it will 'burn itself out' very quickly? How quickly would it shutdown once you stop adding fuel - I guess a second or two?
Please continue from here. Properly cite the original writer in any quotes.

Posts: 1438
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

a few things are required for fusion in a polywell - well, in anything -

1) very high kinetic energy of nuclei
2) very high density of nuclei.
3) very high purity (interfering nuclei that won't fuse will dilute / obstruct)
4) ionization

for this, in polywell, 2 things of note are strong requirements:

1) very strong magnetic fields
2) very high vacuum

and esp. 1).
the moment the mag fields shut down, the central core can be compared to a helium balloon in outer space. it will pop in under a microsecond. and the fusion rate - the density squared part - will reduce faster than the square of that speed.

furthermore, the air in this balloon is electrically charged - meaning it would repel itself. so it would actually considerably faster than a helium balloon in space.

now bear in mind, also, that its the electrons being held in the center by the mag fields that are giving the ions their kinetic energy. and those electrons are MUCH lighter than the ions, so as soon as the mag field falls they are going to explode at near the speed of light, (as fast as a high voltage travels down a near frictionless wire), and with them goes the acceleration gradient for the ions. they will immediately slow down any ions traveling toward the center, and in any case that is their last pass - once they hit the center, they will be traveling outward, and they will not be turned back in.

turn off the vaccum, well i needn't go into that. i think u already have a clear enough picture of what happens when this thing shuts down. the fusion stops orders of magnitude faster than you can pop a balloon in space.

or, if you consider the ion kinetics, the reactor will be completely shut down in less time that it takes a 100kev ion to travel half a meter.

which, needless to say, is PRETTY darn FAST!

mind you if the mag field is even driven slightly out of whack the fusion energy will drop sharply. even a partial mechanical failure would result in a swift and decisive attenuation of the fusion power, if not a complete shut-off.

Posts: 106
Joined: Wed Nov 18, 2009 7:33 pm

Post by jsbiff »


That answers at least one of the questions - as I thought, fusion shut down would be, for all practical purposes, instantaneous, but it still leaves a few questions open: are there any radioactive fusion products in the chamber, and if so, how much, and would the residiual thermal energy of the plasma present any risks, and how quickly would the residual thermal energy dissipate?

Posts: 573
Joined: Mon Oct 15, 2007 4:14 pm
Location: Portland, Oregon, USA

Post by kurt9 »

If a fusion reactor shuts down, the shutdown will be very quick as it takes a maintained vacuum and ion sources to sustain the fusion reaction. There will be very little decay heat. The big problem with D-T fusion is the neutron flux, which is much higher than fission per amount of energy produced. This, of course, is the reason for the focus on developed advanced fuel fusion such as pB11 or He3. The other technical issue with D-T fusion is the need to breed Tritium. It seems to me that if we are stuck with D-T fusion, it make sense to make it a part of a fusion/fission hybrid reactor.

It seems to me that if advance fuel fusion is impossible, that Gen IV fission concepts such as LFTR, IFR, and traveling wave reactors are the way to go. Traveling wave reactors, fueled by waste and depleted Uranium, will give us 500 to 1000 years time to develop fusion power.

Posts: 6339
Joined: Sun Sep 28, 2008 2:29 pm

Post by Skipjack »

Kurt, what do you think about fusion- fission hybrids such as the ones proposed by Helion? I would think that they could be quite save, but I am not so sure about the decay products with this type of reactor. Do you think that they could have caused simillar problems to Fukushima?
I think that since they are able to burn nuclear waste and burn nuclear material more completely (at least to my understanding), there would probably be a lot less radioactive material on site. Of course the question is how much that would be mitigated by the fact that there would be many parts of the reactor that would be quite irradiated by fast neutrons...
The concept by Helion is at least much less complex than a Tok, so AFAIK, there are less parts that need to be replaced (and that would be easier also).

Posts: 1200
Joined: Mon Jun 30, 2008 4:36 am
Location: 92111

Post by Aero »

Since a fusion plant will shut down nearly instantaneously, what will be the reaction from the connected grid? Will there be significant back emf and will there be time to switch spinning reserves online.

Posts: 1438
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

Aero wrote:Since a fusion plant will shut down nearly instantaneously, what will be the reaction from the connected grid? Will there be significant back emf and will there be time to switch spinning reserves online.
good question.

in any case a power blackout is much less of a problem than a nuclear meltdown would be.

i think for such quick and massive power supply changes, you'd need an advanced form of grid power buffering to act as "first responders", so to speak. something that could be supplied by e.g. flywheel energy storage ( ) or superconducting batteries (mentioned in the theory forum, i believe).

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Fusion produces radiation, it is just that the amounts and nature are different. In a large tokamaks the energy contained in the superconducting magnets is also considerable. I recall that if quenched the magnets could release energy equivalent to several hundred tons of TNT. Presumably this energy could be bleed off with carefully designed and roboust current dumps, but it would be a major concern for the engineers, and would still possibly destroy the magnet. Smaller machines like Polywells or FRC may have similar strength superconducting magnets, but they will be much smaller and thus contain much less energy.

I have heard that a production Tokamak might contain several seconds worth of fuel, so fusion fuel starvation would not be instantanious, but no more than a few seconds. Polywells , etc would be on the order of milliseconds or less.
The fusion fuel also makes a difference. D-T reactions produce high energy neutrons, each of which has plenty of energy to transmute multiple atoms. I have heard that D-D fusion produces ~ the same neutron fluxes compared to fission. . I'm not sure if this referrs to the number of neutrons, or the net energy of the neutrons. The neutrons will transmute various elements in the walls. With carfull selection of materials this problem can be mitigated. Again, I have heard that a D-D Polywell building could be entered within ~ 2 days after shut down. The radioactive isotopes produced by neutron bombardment have short halflives so they dissapear much faster then those associated with fission. Of course that means that if you entered 10 minutes after shutdown you would receive a proportionatly larger dose.
Aneutronic fusion would ease these concerns, especially P-B11 fusion which may have relatively tiny neutron fluxes ( to such an extent that the rare 1/10,000 reactions that produce a gamma ray is the dominate radiation risk that has to be managed). I have not heard that any of these secondary radioactive isotopes would result in residual decay heat that would need active cooling beyond what is included with the active cooling when the plant is operating.

The fusion plasmas are indeed not as strong in transferring heat to structures as their raw temperature would suggest. A Tokamak plasma might be at ~ 50,000,000 degrees C. But at a density of ~ 0.00001 atm the heat content would be equivalent to the same volume of gas at one atmosphere at a temperature of ~ 500 degrees. This doesn't mean that the plasma would not quickly heat the surface layers of a block of metal, but for heating the entire block to some equilibrium temperature, the comparison should hold.
A Polywell with a density of perhaps 0.001 atm and at a temperature of ~ 1,000,000,000 degrees C would be equivalent to a gas at atmospheric pressure of ~ 1,000,000 degrees. This is a significantly greater energy content per unit volume. But remember that the volumes are much less than in a Tokamak. (Nebel mentioned ~ 60,000 smaller volumes to produce the same amount of fusion energy). So, the total plasma heat contained in a Polywell would be 1,000,000 / (500 * 60,000) = ~ 0.03 times that of a Tokamak. I'm not sure the 60,000 fusion density claimed in the Polywell will be obtained, but it illustrates that you have to consider the temperature, the density, and the volume when considering the heat loads (and fusion power). You then have to consider how quickly you can cool the structures surrounding the plasma volume. Also, you have to figure how the plasma/ escaping high temperature particles interact with the wall. Direct conversion would decrease the thermal loads on the walls. There are other considerations that complicate the picture further.That is why the performance scaling predictions for the Polywell may not determine the smallest practical size. The heat loading per square meter of the surface area may be the limiting factor. IE: Even if you can increase the Magnetic field strength to ridiculous levels, the size would still be limited by thermal loading concerns.

Dan Tibbets
To error is human... and I'm very human.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Speaking of residual decay heating and neutron induced radioactive isotopes and the thermal heat they would then put out might seem confusing when
I claim that this secondary heat decay is not a problem in fusion reactors, when the neutron fluxes may be similar in both fusion and fission reactors. If so, remember that the decay heat that keep fission fuel rods hot is not due to to neutron induced radioactivity. It is due to the daughter products from the initial fission products, their numbers and the short half lives of some of them. I don't know the details, but apparently there are not too many very short half life isotopes produced from the neutrons bombarding the materials in a fusion reactor wall. Some of the most irrediated parts of a decommissioned D-T or D-D fusion reactor might need to be stored for several hundred years. This "intermediate half life" radioactive waste is convenient as it does not produce enough heat/ unit of time to be a concern, and it does not contain isotopes that need to be safely stored for hundreds of thousands of years.

I wonder if a fusion/ fision hybrid reactor burning thorium, through Ur 233, has a significantly different decay heat production profile.

Dan Tibbets
To error is human... and I'm very human.

Posts: 1200
Joined: Mon Jun 30, 2008 4:36 am
Location: 92111

Post by Aero »

Regarding the direct conversion concept and emergency shutdown --- Would it be possible or practical to feed the energy stored in the magnets into the grid? Would there be enough energy to provide a graceful shutdown of the reactor (WRT the grid) allowing graceful switching to back up reserves? And is there confidence that the this magnetic energy would be available for this purpose?

H mm - I guess I'm also asking what the expected failure modes of a full up, direct conversion BFR would be. Cooling failure, magnet failure (one) converter degradation, failure. Others?

Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »


the answer to your question on the comparison of dangers from fission to fusion waste is actually a question of chemistry.

When U-235 fissions, it splits into binary pairs of isotopes. These isotopes range in atomic mass from around 70 to 160. This means that in a used, depleted reactor rod, you will have every chemical element in the middle of the periodic table. This is a toxic chemistry set of near-unlimited potential, particularly if you then add in some hydrogen and oxygen from a cooling system!!

The distribution of isotope masses from U-235 fission has a distinctive double hump, with isotopes of masses ~95 and ~135 being more prevalent that other atomic masses.

We find that amongst the jungle of atomic species around 135 mass that the ones that stick out - toxic chemistry wise - are isotopes of iodine, caesium. In the ~95 masses we find strontium and technetium.

Also, you will note Xenon and Krypton fit in here, and such isotopes [due to their lack of chemistry reactivity] will simply drift off to be breathed in somewhere.

There are now two scenarios of waste risk to consider - short term and long term. For long term risks, once this stuff is locked up in a sealed cask and buried in a thick granite seam 1 mile underground, the reality is that the only risks are to political reputations.

But there are serious risks from the short term waste of spent rods, as you are currently seeing in Japan: Iodine will readily vaporise once hot and drift away, and caesium oxidises very energetically, and may also further cobine with iodine to make CsI aerosol. These are particularly dangerous because they are a) the most common emissions from 'distressed' spend rods and b) are very active in the body, with Cs displacing any K and with I being soaked up readily by the thyroid gland.

These two isotopes are quite short lived, so you'd not find them in 1000 year old spend rods, but their short life also means they are more radioactive. Isotopes with short lives means they disintegrate more readily, so are far more radioactive, per mass, that the likes of U that takes billions of years.

I would suggest [I'm not a 'nuclear heath' person, so this is a bit of speculation/suggestion of my own] there is a simple way to look at this chemical+nuclear risk; if you can protect yourself against these two insidious and high radioactive isotopes, then you are already over the worst and the others in the menu of nasties are 'relatively benign' by comparison with how readily these are taken up by the body. But if you can't protect yourself from Cs and I, then the other nasties are the least of your problems.

Now we turn to fusion, which I will close by simply saying the rather obvious statement that no-one has yet build a working reactor so we can't yet know what types of materials will be essential for its operation. But what we can say is that we won't have every chemical element in the periodic table floating around inside the reactor, unlike a fission reactor, so in that sense any radioactive waste can be 'designed for' and will arise in a controlled manner by selection of reactor materials (which will be chosen by their low rate of absorption by humans, and their low activation cross-sections). So, no nasty Cs and I

Fusion is also very difficult to do and for thermonuclear plasma devices, like tokamak, the plasma could easily be [deliberately] disrupted, killing the reaction in microseconds. For other devices, many are pulsed machines anyway [Polywell has not been operated for any longer than a few milliseconds] so it would simply be a case of not powering it for its next pulse.

Posts: 2261
Joined: Fri Oct 26, 2007 8:50 pm

Post by hanelyp »

For a fusion reactor we have only a few likely fuels to choose from, most of which produce lots of neutrons. But we can choose the inner walls of the reactor for low neutron activation. Boron, Carbon, and Silicon are all decent on that front.

Posts: 256
Joined: Thu Aug 28, 2008 5:04 pm

Post by pfrit »

The big difference between the neutrons produced in a fission reactor and a fusion reactor is where they would be absorbed. In a fuel rod the neutrons are in a dense material of heavy atoms, so most of neutrons (ok, many) will be absorbed by the surrounding fuel. Thus the chain reaction. In a fusion reactor, the neutrons would be in a vacuum and fly out of the reactor without hinderance. Thus, a higher neutron flux. And aneutronic fusion is a misnomer. The side reactions will produce many neutrons. I know this has been heavily discussed. Now I don't think that this would be a problem except in how the SC magnets are effected. We do know how to build things in a neutron flux.
As far as the magnets, quenching is a major issue and they would be a major part of the expense in building a polywell. And they will fail. The flux demands that they will. Earthquakes and tsunamis will do them in as well. And when they quench, they will explode. Pehaps not doing much damage to anything besides the magnets themselves (I kinda doubt that), but the magnets will be destroyed. Expensively. The actual fusion waste products are meaningless in terms of danger, but the waste at decommisioning could be considerable.
All sums up to expensive, but not dangerous
What is the difference between ignorance and apathy? I don't know and I don't care.

Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

pfrit wrote: The actual fusion waste products are meaningless in terms of danger, but the waste at decommisioning could be considerable.
That's not entirely true. biff had it right discussing tritium. As it is hydrogen, it is readily absorbed into materials and recovering and maintaining traceability of the tritium that's gone into the reactor is a Big Deal, and it is a fusion product as well in various fusion branches of different reactions. The 10 year half life makes it 'significantly' radioactive, but it will hang around in the environment for a long while. But in general, if you can get most of the fusion product out as 4He then, of course, that is fine.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

The neutron effects on structure, superconducting magnets, radioactive isotopes is complicated and not fully explored. And, neutron production and it's effects are absolutely nessisary for certain types of fusion. In Tokamaks, the only possible fuel is deuterium and tritium, and the tritium has to pe produced through neutron bombardment as it does not occur naturally (except in tiny trace amounts) due to it's relatively short half life.
Even in a reactor that might operate without tritium you might wish to utilize it as it is a byproduct of D-D fusion and you have to handle it anywhay. Depending on the efficiencies, a Polywell may need this "D-D 1/2 catalized" process. Bussard mentions this as one option. D-D fusion results in tritium and Helium 3 production. These can be collected, processed and feed back into the reactor as new fuel, along with the Deuterium. The tritium in peticular would increase the net fusion rate and subsequent energy output.
With D-D of D-T fusion, the neutrons may be a desired side product as they could generate additional tritiums in a lithium blanket, or in a Boron 10 blanket to produce more heat and Helium 4 as a decay product (?).

With widespread D-D Polywells used for grid power, the Helium 3 could be stored for use in ships or spaceships where the neutron flux needs to be minimized as much as possible (and where you wish to maximize the direct conversion potential). This route to Helium 3 production might be even more economical than trying to mine it on the Moon. P-B11 fusion may have large advantages over D-He3 fusion, but it has it's own set of problems.

Dan Tibbets
To error is human... and I'm very human.

Post Reply