There is another way of doing fission, that was demonstrated (working prototype and all) in the 60's: molten salt. You basically dissolve the fissile matter in molten salt, provide initial heat, and you have a reaction going. This has multiple advantages:
- Originally designed for a nuclear plane, it is capable of reacting to load changes very quickly. It's basically easy to power-up, and easy to shut down.
- It is amenable to a passive security set up: when all goes wrong, the reactor shuts down by default. When it does, the radioactive matter is trapped in no-longer-molten salt, such that contamination is limited even if there is a leak.
- It consumes 90% of the fissile matter before it has to go to recycling. (Current reactors do about 10%.)
- It currently produces 5 times less waste than current reactors. There is hope to do even better.
- Last but not least, it can run off thorium, which is 4 times more abundant than uranium, and much better spread out geographically.
With one or two decades and a few billion dollars, we could make commercial grade reactors. They would almost certainly be significantly cleaner, and much safer than current light water reactors. They could adapt to changes of demand in the power grid, or compensate for the volatility of most renewable power sources without the need for a "battery miracle". Overall, they would make a terrific mid-term solution, that could last a couple centuries. More than enough time to come up with something better.
I'm sure with some extra R&D people could make those reactors reliable, but if you search the new generation reactors on Wikipedia, you'll see every one of them ended in some kind of failure.
Another possible cause is path dependence. We have a whole industry around solid fuel, light water reactors. Switching to molten salt throws much of that away.
Much research still need to be done. We may want cleaner salts, we want even less radioactive waste, and we definitely want to accurately assess the safety of this stuff ("passive security" is not enough by itself). This means more prototypes reactors, of various sizes, each more expensive than before. We want a few million dollars for the small ones, then a few billions for the big ones.
It may not be just lack of founding. Materials science has advanced a lot since then, and those problems look solvable now, but I'm not sure they could be solved by the 60's.
To quote "Technologies Underlying Weapons of Mass Destruction",
Reactor-grade plutonium recovered from civilian reactors differs from weapon-grade plutonium in the relative proportions of various plutonium isotopes. Reactor-grade plutonium has a higher rate of spontaneous fission reactions than weapon-grade, generating neutrons that can initiate the nuclear chain reaction during weapon detonation sooner than would be optimal. As a result, using reactor-grade plutonium in a first generation nuclear weapon can significantly reduce both the predictability and the expected yield of a weapon designed by a proliferant state.
None of the states that have either made nuclear weapons or attempted to do so appear to have selected anything but high-quality plutonium or uranium for their designs. Nevertheless, from a technical perspective, reactor grade plutonium can be used to make nuclear weapons, and any state possessing significant quantities of separated plutonium should be considered to have the material needed to fabricate nuclear components for nuclear explosive devices in a short period of time.
http://www.au.af.mil/au/awc/awcgate/ota/9344.pdf
If you want to seize on the part where it says that one could build bombs from civilian reactor waste even though it's less straightforward and no nuclear weapons state has ever done that, then I'll point out that the thorium fuel cycle likewise could be used for bombs even though no NWS has ever taken that path (Quoting LANL's "Thorium Based Power Systems and Relevant Safeguards Considerations"):
Thorium is not safeguarded until it is converted into fuel materials. It would seem that if the starting point for uranium safeguards is moved up and thorium begins to play a larger role in the global nuclear establishment, then thorium safeguards should move to the equivalent point in the thorium fuel cycle.
Unlike thorium, U-233 is subject to stringent safeguards. An SQ of U-233 is only 8 Kg, the same as plutonium. U-233 is a highly desirable weapons material despite any practical U-232 content. According to the material attractiveness studies done at LANL, PNNL and LLNL, the additional dose from U-232 decay products is insufficient to stop a determined adversary. While it may be somewhat beneficial to physical protection, the dose rate is not high enough to incapacitate workers. Dose rates can be greatly reduced with shielding and military personnel handling a fabricated weapon should be able to have reasonable dose rates using established techniques. For example, the pit can be stored in lead shielding and only placed in the weapon immediately before use.
http://permalink.lanl.gov/object/tr?what=info:lanl-repo/lare...
The actual military lineage of civilian LWRs is that the US Navy developed pressurized water reactors fueled with enriched uranium to power vessels in the 1950s, starting with USS Nautilus. That variety of reactor had the fewest unknowns when the civilian nuclear industry first developed in the US and the path dependency effects continue today.
However, this was probably not the case for the first reactors. If I recall my documentary properly "la face gâchée du nucléaire" (French and German only I'm afraid), the technology was initially chosen because it could produce plutonium. Therefore it got lots of research, and naturally ended up working best.
Submarines were of course responsible for much of the continued funding. Molten salt got cut off in no small part because the very idea of a nuclear plane was not very good.
That said, I'll need to study this subject more closely before I consider myself properly informed.
The very first reactors were indeed built for plutonium production. Those 1940s reactors built in Hanford, WA, USA used unenriched natural uranium for fissile material, used purified graphite as their neutron moderator, and did not generate any useful energy. In fact they were net energy consumers because they needed an external power source to move water through to keep them cooled.
13 years after the Hanford reactors started, Shippingport became the first civil power reactor in the US: https://en.wikipedia.org/wiki/Shippingport_Atomic_Power_Stat...
Shippingport literally started with a surplus naval reactor from a cancelled aircraft carrier project. It was a pressurized light water reactor requiring enriched fuel like most reactors now operating commercially. Everything about it was different from the early plutonium reactors: the fuel, the moderator, the heat exchange... by a funny coincidence, Shippingport was also the only commercial reactor in the US to use the thorium fuel cycle, from 1977 to 1982.
Maybe the misunderstanding started from the fact that the first British power-generation reactors were intended to produce weapons-grade plutonium. Those reactors were quite different from light-water reactors: graphite-moderated, gas-cooled, and using unenriched fuel.
The only up side is that the decay products have a shorter half life. So the waste will only be a problem for a century or so, instead of for millennia.
This may be better than fission, but it's definitely not clean by any realistic definition of the word.
By contrast the byproducts of a fission reaction are set by the fuels used and include really nasty stuff like Strontium-90 and so on. But even fission plants are designed so that the radioactive byproducts are essentially all spent fuel or things contaminated by spent fuel. Neutron activation is a design consideration rather than a source of waste.
Of course the neutron flux will tend to enbrittle the metals used in it's construction which will have the effect of increasing maintenance costs.
> The only up side is that the decay products have a shorter half life. So the waste will only be a problem for a century or so, instead of for millennia.
We have people who live 'for a century or so' but we don't even have languages (now!) that have lasted a millenia.
Wut?
Also, Tibetan maintains the same spellings from 1200 years ago, but that really just means that it's very, very difficult to spell, since the pronunciation no longer matches spelling well.
These make interesting exceptions to the general rule, so I think the original point is still very valid. I'm curious about old vs. modern Syrian, if anyone knows about that.
I'd posit that storage is not a solved problem, and a big part of that is a lack of confidence that we can reliably communicate with the people who will have to deal with this waste in the distant future.
There's also the inherent disdain involved in passing on massive costs and risks to future generations to solve a comparatively short-term problem of ours.
Most importantly they are vastly safer and thus need far less security and far less insurance. Remember a huge benefit for PV solar and wind is they don't need guards.
PS: Another consideration is the half life is short enough that over 20 years things become a lot less radioactive. Fission has some half lives just long enough to be a real pain.