6

Setting

The full society collapse is starting, and a group of people know it. It may or may not lead to a total extinction event. They can't protect a whole city but a couple of hectares is doable. The group is less than a hundred people. There is no real intent to protect/restore humanity as a whole, at least not until the apocalypse will end. It's a very near future, where reasonable tech advances of today are not experiments anymore, but commercial solutions instead. Fusion is still 20 years in the future, though.

Premise

They capture a fast breeder nuclear power plant which is operating and already manned by the qualified personnel. This type of the power plant is a common one in their timeframe, and watercooled "traditional" ones are slowly being obsoleted. The nation power grid is still functional, refueling happened yesterday and the nearby city is not razed to the ground yet. The nation is a powerful nuclear one, there's plenty of resources to plunder in future.

Problem

  • What would be the immediate problems to solve?
  • What would be the short-term problems to solve?
  • What would be the long-term problems to solve?
  • What minimal set of issues should I handwave?

Handwave food, aging and dangers to the power plant itself. Assume absolute protection of the facilities. Don't handwave water. Clean drinking water is still important.

Notes

  1. This is different from the Can Average Joe reboot the nuclear power plant and Running a nuclear power plant in the post-apocalypse: is it possible? because the power plant is already fully functional and manned by the qualified personnel.
  2. Judging from the How useful would a hydroelectric plant be in a post-apocalypse-world it's doable with the dams. But for several reasons it is not desirable to replace a nuclear plant with a hydro one in the story.
  3. Fast breeder reactor in general was chosen to reduce the need in the fuel. I understand that there's a lot of other problems aside from the fuel to think about. :D
  4. Thank you.
hijarian
  • 257
  • 1
  • 8
  • 2
    Minimum support for a highly experimental nuclear reactor is probably 1000+ workers. Full-time. Overtime. They need all sorts of high-tech instrumentation that requires periodic replacement, etc. They need nation-state levels of expertise, probably a functioning university system. Able to spend billions worth (or its equivalent) of revenue. Given the constraints of your question, this plant can only run 3-6 months before something inevitably makes it no longer possible to safely run. They have a moderate chance of a controlled shutdown. – John O Feb 09 '21 at 17:30
  • @JohnO Thank you for noticing that! I'll add the note in the text. Assume that it's a near future, and this power plant is not an experimental anymore, but a standard one (as standard as a nuclear power plant can be). – hijarian Feb 09 '21 at 17:44
  • 1
    The staff will be very stressed. Reactors have stringent testing and surveillance requirements and without outside support - new parts, NRC inspectors, etc - the reactor will be mostly fine but the staff will be unhappy about it – Sol Feb 09 '21 at 18:35
  • 1
    It can't be maintained, regardless of whether it's nuclear, hydro, or fossil power. It faces the same problem as any complex machinery in a post-apocalyptic scenario: no replacement parts for things that break or wear out and no supply consumables like lubricants, etc. And there's no way to jury-rig replacement bearings or gearboxes of the size involved. The same issue applies to the power grid that the reactor is attached to. – GrumpyYoungMan Feb 09 '21 at 19:21
  • 1
    How many square miles of spare parts and tools warehouses do you have access to? You will need spares of, literally, millions of different parts. From duct tape to #4179screw(nonmetallic,radiationresistant) to toothpaste to turbine blade(replacement, type 17b/477 do not confuse with turbine blade replacement 17b/447)., etc, – PcMan Feb 09 '21 at 22:39

1 Answers1

6

Building a breeder reactor is easy, but not a fast breeder

Building and operating a breeder reactor on post apocalyptic technology is the easy part. It is almost trivial to build a graphite breeder reactor that uses a thermal neutron spectrum with natural uranium. If you want a fast breeder, however, you need enriched uranium, something that will almost certainly be beyond your post-apocalyptic tech. This reactor would breed fissile plutonium, which could then be reburned in your reactor.

Reprocessing the fuel rods into new fuel is the hard part

The reprocessing of spent fuel requires enormous industrial infrastructure. The first problem is that you're going to need an industrial-sized hotbox for working with the spent fuel rods. This hotbox needs remote controlled robotics, video cameras, etc., and other technologies that will be difficult to come by in a post apocalyptic world.

The even bigger problem is that you will need industrial scale chemical engineering to properly separate out the fuel from the wastes and neutron poisons.

Add to this that you can't reprocess the fuel immediately. The irradiated fuel in a breeder reactor fuel cycle often needs to cool for at least 10 years in a spent fuel pool before it can be safely handled and worked with. Even then, it would be immediately lethal to anyone within 10 meters of an uncovered fuel rod, hence the requirement for a strong hotbox.

But why though?

Ultimately, the breeder reactor concept fell out of favor for one main reason: Uranium is plentiful. It's far cheaper and easier to dig up new uranium and burn that than to try and deal with all the mess of a breeder reactor. The infrastructure required to make new natural uranium fuel is far less complex than what you'd need to process the fuel rods from a breeder reactor.

Even reprocessing spent fuel from a non-breeder reactor is more economically feasible.

Other options

There are a number of exotic reactor designs on the table right now that, at least in theory, are safer and easier to operate than the PWR and BWR designs currently used in the American nuclear reactor fleet.

The Liquid-fluoride Thorium Reactor (LFTR) is probably the best design. It's a breeder reactor, is impossible to melt down because its fuel is already molten, can't explode because it doesn't use materials that can produce hydrogen, and works with thorium, which is far more common than uranium. It also can't be driven into a dangerous prompt critical state because it is self-limiting.

Such a reactor would be far more likely to be operable by the average apocalypse survivor, but unfortunately it would be far less likely to be maintainable.

Much like a breeder reactor, a LFTR reactor requires significant chemical infrastructure to properly filter out fuel from the molten salt. So even though the reactor is so safe and easy to use that someone with barely a high school education could operate it, they wouldn't be able to run it for very long since they'd need a literal chemical processing plant to keep it refueled.

stix
  • 5,913
  • 1
  • 17
  • 30
  • Thank you for your answer. What timeframe for LFTR refueling we are talking about? 3 months before refueling, 1 year, 10 years? – hijarian Feb 09 '21 at 17:56
  • 1
    @hijarian Less than three months. Of course, the reactor could have a good quantity of fuel on-site. LFTR (hypothetically) are almost continually refueled, because the reactor does not have to be (and should not be) shut down to do so. – jdunlop Feb 09 '21 at 18:19
  • 1
    @hijarian Since one hasn't been built, there's no saying. Most designs involve continuously filtering the coolant salt and then slowly adding back the bred U-233. It's more a problem of filtering out impurities than refueling. The reactor itself will run for awhile so long as the coolant salt is kept clean. – stix Feb 09 '21 at 18:19
  • 1
    travelling wave reactor should be interesting. it doesn't need any refueling. – ths Feb 09 '21 at 20:58
  • 1
    @ths A TWR has no refueling because it's essentially a "disposable" reactor that has no user servicable parts. Once it burns up its fuel you replace it with another TWR. That doesn't really fit with the OP's spirit of a breeder reactor. – stix Feb 09 '21 at 21:00
  • 1
    yeah but it lasts for decades. – ths Feb 09 '21 at 21:02
  • 1
    @Mazura that is not entirely true. The reason we don't reprocess fuel is purely political and related to nonproliferation. You don't need a breeder reactor to reprocess spent fuel. Economics and engineering are the reason we don't build fast breeder reactors. It simply isn't cost effective to do so. In addition, the nonproliferation concern is really a red herring, since it is easier to build a military reactor specifically to produce plutonium than it is to divert waste from a civilian reactor. A weapons reactor for plutonium needs no enriched uranium and no special engineering. – stix Feb 10 '21 at 17:46
  • @ths this is a nice suggestion btw, as decades is a timeframe I can work around. Thank you - first time I hear about the TWR design! – hijarian Feb 19 '21 at 19:19