Lightest Fissile Element/Isotope/Nuclide?

[sanman]

So I'd like to know - what is the lightest element/isotope/nuclide that could be feasibly/practically used in a nuclear fission reactor for energy production?

Has any research been done into this?

What is the lightest element that's actually been used in a controlled sustained fission reaction?

[Azael]

uranium-233 is the lowest that can and has been/is used. The isotopes of elements below uranium has very small fission cross sections.

[sanman]

And yet there are nuclides below that weight which are unstable, and have a high rate of decay. Why would a nuclide slightly below U-233 be unable to sustain a fission reaction, if it was compressed to suitably high density?

If U-238 can undergo an explosive chain reaction when compressed to supercritical density, then why can't a nuclide slightly below it even sustain a fission chain reaction if sufficiently compressed? Why would the nuclear cross-section be so much lower in that case, that a chain reaction would not be possible even under heavy compression?

[Astronuc]

And yet there are nuclides below that weight which are unstable, and have a high rate of decay. Why would a nuclide slightly below U-233 be unable to sustain a fission reaction, if it was compressed to suitably high density?

If U-238 can undergo an explosive chain reaction when compressed to supercritical density, then why can't a nuclide slightly below it even sustain a fission chain reaction if sufficiently compressed? Why would the nuclear cross-section be so much lower in that case, that a chain reaction would not be possible even under heavy compression? By compressing to a suitably high density, does one mean compressing as in an implosion?

Compression does not affect the microscopic cross-section. That is an inherent property of the nucleus. The compression causes an increase in density, and the affects the macroscopic cross-section, which is proportional to the atomic density of the material.

In fact, U-238 fission more readily by fast neutrons, not thermal neutrons, and uranium fissile systems are enriched in U-235, which is more readily fissionable by thermal neutrons, as is U-233 and Pu-239.

The less heavy elements between Bi and Pa prefer to decay by beta or alpha emisson, rather than fission.

[sanman]

But even by increasing the bulk density, we are increasing the probability density of the neutrons traveling through that same space. And for a given nuclear cross-sectional area, more neutrons per unit space means more likelihood of collisions.

Furthermore, couldn't we provide some kind of seed trigger, by some initial dose of neutrons, to get more fissions happening? If that is timed with the implosion to achieve a sufficiently high density, then why can't we tilt the odds in favor of a chain reaction?

[Astronuc]

One could provide a seed of neutrons or neutron source, but that is not the issue.

The issue is that increased atomic density doesn't increase the number of neutrons, on the probability that any neutron in the material will react with a nucleus before it escapes.

Implosion are short term transients. The imploded material pushes back as it heats, and expands as applied pressure decreases.

Finally, even if neutrons are absorbed, the absorbing nuclei are more likely to undergo an (n, gamma) reaction in which the A+1 radionuclide de-excites by gamma emission, and does not undergo fission if it is a non-fissile nuclide.

[sanman]

Implosions could be repeated cyclically. The point is to get out more energy than you put in, just like with a piston engine.

Or even without implosions, it might be possible to compress a material very densely:

http://www.sciencedaily.com/releases/2008/03/080320095005.htm

(I also want to discuss this linked article in connection with muon-catalyzed fusion, but I'll do that in a separate thread)

Anyway, so suppose some nuclide not far below uranium could be suitably compressed and kept at high density for a sustained period. Then, regarding whether or not that nuclide would tend to absorb neutrons vs fissioning, would be determined by the stability of that nuclide. So just pick a nuclide that's suitably unstable, and therefore it would be more likely to break than to stay together while de-exciting by gamma-emission.

[vanesch]

I can be wrong, but I think that the reason that you cannot make a critical mass (Astronuc referred to this) with pure U-238 is the following. The total cross section is higher by a factor of about 5 above the fission cross section in the 1MeV region (where it is in fact the closest). This factor is bigger than the average number of neutrons produced in a fission reaction. As such, with the number of neutrons produced in a fission, you will never produce 1 new fission or more.

In the attachment, the blue curve is the fission cross section while the red one is the total cross section (data from the sigma application on http://www.nndc.bnl.gov/ ).

[sanman]

And yet there are some materials such as Thorium, which are known to be able to undergo fission under a spallation source. They are thus subcritical reaction materials, only able to exhibit a fission reaction with the aid of the external spallator. But that spallation source is then not increasing Thorium's nuclear cross-section, but again only simply increasing the neutron flux in the material.

So for a material like Thorium then, couldn't a similar increase in neutron flux be achieved by increasing the density of the material through compression?
Even if we're only talking about a brief compression, like an implosion, could it not then achieve a sustained fission chain reaction for that brief duration?

Or if there was some way to apply a sustained high compression, such as with the buckyball confinement example I gave, then could this not improve the neutron flux enough for a sustained chain reaction?

[Astronuc]

So for a material like Thorium then, couldn't a similar increase in neutron flux be achieved by increasing the density of the material through compression? A spallation source is an external neutron source which has to added to the thorium to get a neutron flux. Thorium fueled reactors need U-233 or U-235, but the Th-232 does capture an neutron becoming Th-233 which eventually transforms to Pa-233 which transforms to U-233, and thus one can do a thermal breeder reactor as opposed to a fast breeder.

The point of accelerator driven systems is that they put a lot of energy in from the outside, as opposed to within the fissile system, and IIRC the point is to make them subcritical.

Even if we're only talking about a brief compression, like an implosion, could it not then achieve a sustained fission chain reaction for that brief duration? If one had a mass of Thorium surrounding a neutron source, e.g. highly enriched fissile kernel, the an implosion might work, but the reaction is very brief, i.e. explosively so, and that is not a practical energy source. The idea of a chain reaction is a sustained and well-controlled 'steady-state' process - not a pulsed detonation.

Or if there was some way to apply a sustained high compression, such as with the buckyball confinement example I gave, then could this not improve the neutron flux enough for a sustained chain reaction? One still needs a substantial mass of fissile material to maintain a steady and well-controlled chain reaction, and the buckyballs would be destroyed by the fissions.

[sanman]

A spallation source is an external neutron source which has to added to the thorium to get a neutron flux. Thorium fueled reactors need U-233 or U-235, but the Th-232 does capture an neutron becoming Th-233 which eventually transforms to Pa-233 which transforms to U-233, and thus one can do a thermal breeder reactor as opposed to a fast breeder.

The point of accelerator driven systems is that they put a lot of energy in from the outside, as opposed to within the fissile system, and IIRC the point is to make them subcritical.

If one had a mass of Thorium surrounding a neutron source, e.g. highly enriched fissile kernel, the an implosion might work, but the reaction is very brief, i.e. explosively so, and that is not a practical energy source. The idea of a chain reaction is a sustained and well-controlled 'steady-state' process - not a pulsed detonation.

Well, the idea of burning a candle is a steady combustion. But the idea of a piston engine is a pulsed combustion. Both can be harnessed for useful purposes.
It's possible to imagine a cyclical pulsed process, which briefly achieves high compression, over and over again.

One still needs a substantial mass of fissile material to maintain a steady and well-controlled chain reaction, and the buckyballs would be destroyed by the fissions.

Perhaps there could be some kind of moderator material in between the buckyballs and the thorium inside them.

Or alternatively instead, the buckyballs could be continuously "healed"/repaired by electromigration under strong electrical current. Or perhaps some other high-temperature carbonaceous/hydrocarbon material could be flowing around the buckyballs' exterior.
Buckyballs can themselves be formed from a feedstock of ethylene gas at high temperature, can they not?

[Astronuc]

Well, the idea of burning a candle is a steady combustion. But the idea of a piston engine is a pulsed combustion. Both can be harnessed for useful purposes.
It's possible to imagine a cyclical pulsed process, which briefly achieves high compression, over and over again. Try - piston combustion engines work on a cycle detonation process - but they are relatively small scale - and more importantly - they don't produce fission products!

Perhaps there could be some kind of moderator material in between the buckyballs and the thorium inside them.

Or alternatively instead, the buckyballs could be continuously "healed"/repaired by electromigration under strong electrical current. Or perhaps some other high-temperature carbonaceous/hydrocarbon material could be flowing around the buckyballs' exterior.
Buckyballs can themselves be formed from a feedstock of ethylene gas at high temperature, can they not? What is the size and structure of a buckyball? Then think about what a 50-80 MeV atom (fission product) would to do the atoms in a buckyball. A strong electric current in a large reactor is going to use a lot of energy, and I'm quite sure it won't do much to repair buckyballs in a high radiation environment.

[vanesch]

And yet there are some materials such as Thorium, which are known to be able to undergo fission under a spallation source. They are thus subcritical reaction materials, only able to exhibit a fission reaction with the aid of the external spallator. But that spallation source is then not increasing Thorium's nuclear cross-section, but again only simply increasing the neutron flux in the material.


Every material with a fission cross section can undergo fission (of course) ; the point is that to get any hope of having a net generation of energy, you need a serious amplification of the neutron flux (that is, the neutron flux in the core needs to be a serious factor (say, 20) higher than the incoming neutron flux from the accelerator/spallation. In a normal nuclear reactor, this "amplification factor" is infinite (it is a self-sustained chain reaction), while in a sub-critical reactor, this amplification factor is finite. In a critical reactor, the "k-factor" is 1, while in a subcritical reactor, this is slightly less than one (say, 0.95 for an amplification of 20). But you still need a k-factor very close to 1. And *that* is something that can only be achieved with certain materials.

Now, the k-factor is obtained by considering how many fissions are obtained by the neutrons released by one single fission (you understand that if, for each fission, we cause another fission (k=1), then the reaction is self-sustaining). In the calculation of k, one finds of course:
- the average number of neutrons produced by a fission (A)
- how many of them get lost by other processes, like capture, before causing fission (B)
- how many of them get lost by "geometry" (C).

A is a property of the fission process (and is slightly dependent on the spectrum of the neutrons) ; B is given by the RATIOS of the cross sections and the mixture of different elements. C is given by the size and the density of the material.

Now, if you consider an INFINITE amount of material, then nothing gets geometrically lost, and we have k-infinite. For a specific setup, we have k = k-infinite x g where g is a geometry factor between 0 and 1. If one compresses a material, then one brings g closer to 1. It is the trick one uses in an implosion atomic bomb: k-infinite is of the order of 2, and the g of the non-compressed material is less than 0.5, while the g of the compressed material is close to 0.8, so the k-factor goes from less than 1 (subcritical) to more than 1.6 or bigger (fast divergence).

But if k-infinite is less than 1, no compression can ever achieve criticality.


So for a material like Thorium then, couldn't a similar increase in neutron flux be achieved by increasing the density of the material through compression?


Nope, because for thorium, the k-infinite is less than 1.