...no, I mean, supposedly, I believe there is more about it than mere kans on our part. Installations like this one are found all along the spoorwegs, and yes, around some you should tread lightly. Yet most are safe, and held, maybe still hold, functional warmers! What was th-

Ade, 2nd Class Engineer of Brug Gibraltar before getting shot

Welcome to a future. Mankind has brought doom upon themselves, their cities have been flattened by war and weather, and most of the northern hemisphere is radioactive badlands.

They put their differences aside to fix keep the planet, reaching an unprecedented scale of globalization, making major breakthroughs as a unified people - just to bomb themselves to smithereens once that immediate fear was off the table.

In fact, the fragile peace lasted less than a decade without the external pressure of an extinction event providing something to be worked on.

Once that fate became clear, governments hastily began restoring and expanding abandoned and repurposed cold war era bunkers. To little avail. Only a fraction of the populace managed to find shelter when the bombs started falling.

Months to years, and in some cases decades, later survivors start emerging again. They are quick to start establishing new communities. Unfortunately, the electro-magnetic-fallout from the bombs fried a majority of the world's power- and communication-grids, preventing these places from easily connecting with each other.

Yet, sturdier tech survived relatively unscathed. Most places are still connected by rails and a fair number of pre-war train-engines are found in rail-depots in various states of disrepair - ready for their new job as the lifeline humanity 2.0.

The trains are powered by huge Radioisotope Heater Units (RHU), using Radium, as described in this question on the feasibility of such engines.

For the purposes of my story I am aiming to have RHU storage facilities in rail-depots and/or other secure locations (e.g. tunnels), where RHUs are stored in pits in the floors and hoisted up by overhead cranes to be transported or installed into engines.

I was originally thinking about having them submerged in a heat-conducting liquid and using the heat to drive turbines to power workshops and machinery around the facility. But to my understanding there are at least 2 major issues with this approach considering I need these facilities to last while there are no people to maintain them for up to multiple decades:

  1. Breakdown of moving parts: when (not if) the heat exchangers or secondary cycles start breaking down the heat will accumulate in the main 'pit' and eventually ignite things.

  2. Decay of the submerging fluid: Water would amass algae and other biomatter, oil would degrade over time, car coolants seem to last about 5 years tops - that is nowhere near long enough...

Q: How can these RHUs be securely stored for decades to even centuries?

Solutions will be judged by the following criteria weighted from top to bottom:

  • engineering complexity: The fewer moving parts the better
  • recoverability of RHUs: The less work is necessary to get an RHU back into working condition before it can be reinstalled the better
  • deterioration in storage: The less RHUs deteriorate in storage in regards to time spent in storage the better
  • footprint of installations: The less space is necessary per stored unit the better
  • Availability of used materials: The easier to come by the materials are all around the world the better

This question is about:

  • finding a working storage solution for the RHUs over the time of decades to optimally centuries

This question is not about:

  • what element the RHUs are made of. Please refer to this question if you want to propose alternatives to the described radium RHUs
  • finding processes to generate power from the RHUs, unless it is part of a storage-solution
  • $\begingroup$ @user535733 that is an interesting bit of trivia. Did you just want to share some knowledge or did you mean it as a reaction to some part of the question - e.g. trying to indicate issues with it? E.g. if I am mixing up terminology > why not propose an edit? :) $\endgroup$
    – dot_Sp0T
    Commented Jan 9, 2018 at 13:39
  • $\begingroup$ @user535733 sadly you've deleted your comments without me understanding what you pointed to... Having spent some time looking at the dictionary and pouring over wikipedia it seems that 'railyard' is a term with a very definite meaning in english, which I assumed would include things such as repair-centers. After digging I found the term motive power depot which, I assume, using will remove that factor of misunderstandings. $\endgroup$
    – dot_Sp0T
    Commented Jan 9, 2018 at 15:35
  • $\begingroup$ The details of language don't matter - most folks don't know the difference anyway. If your story is about railways, just make sure your understanding of basic railway operations is correct. Yards, equipment maintenance, and long-term storage are often different departments, and rarely located near each other. The facilities for each have different requirements. $\endgroup$
    – user535733
    Commented Jan 9, 2018 at 20:05
  • $\begingroup$ @user535733 where I live and come from most of these facilites are near each other actually, that's the whole issue isn't it? Also there might well not be a story at all. $\endgroup$
    – dot_Sp0T
    Commented Jan 9, 2018 at 20:08
  • $\begingroup$ Some places indeed have surplus infrastructure with those services clustered together in legacy locations, most notably central and eastern Europe at the moment. New investment (North America, China, India, Africa) generally separates those three services quite widely. $\endgroup$
    – user535733
    Commented Jan 9, 2018 at 20:31

3 Answers 3


Your primary problem is cooling

You will need a constantly cycling cooling system ideally with the ability to automatically top up lost coolant or holding such redundant levels of coolant that moderate losses are not catastrophic.

I would advise using water as your primary coolant. Your concerns about buildup of biological matter can be resolved by the simple consideration of the temperature the system is going to run at. Anything short of extremophiles are going to be cooked off in short order.

Since this is a nuclear facility a consideration for a cooling system that will survive unpowered, unmaintained, and unattended, for example in case of meltdown or earthquake making the building unsafe, could be a part of local regulation for building the yards in that country/state. Based on a scaled up thermosyphon* for example.

While this approach is not entirely bullet proof in the very long run, its largely passive nature and lack of moving parts means that facilities that survive the war should survive a considerable period afterwards.

Facilities will be lost, some will suffer leaks and be unable to recover sufficient water, or suffer earthquake or other damage and suffer catastrophic failure. The nature of a rail network means that while some depots will be lost, others will survive.

Other considerations

Recoverability: This is how they were designed to be stored, this should not be a problem if the facility remained intact.

Deterioration: As for recoverability, hopefully a unit designed to be stored in water will not corrode when stored in water.

Footprint: Well within standard warehouse sizes, as big or small as you like really. I advise making reference to earthquake proofing.

Availability of used materials: Nothing special going on here. Concrete, water, steel.


Thermosiphon (or thermosyphon) is a method of passive heat exchange, based on natural convection, which circulates a fluid without the necessity of a mechanical pump. Thermosiphoning is used for circulation of liquids and volatile gases in heating and cooling applications such as heat pumps, water heaters, boilers and furnaces. Thermosiphoning also occurs across air temperature gradients such as those utilized in a wood fire chimney or solar chimney.

This circulation can either be open-loop, as when the substance in a holding tank is passed in one direction via a heated transfer tube mounted at the bottom of the tank to a distribution point—even one mounted above the originating tank—or it can be a vertical closed-loop circuit with return to the original container. Its purpose is to simplify the transfer of liquid or gas while avoiding the cost and complexity of a conventional pump.


The key to successful RHU storage is the ability to safely separate the Hot Slug (heat source) from the mechanical Power Converter assembly. Each component is likely to have a different design lifetime anyway. Separating the RHU into the two components makes maintenance of the overall system easier.

Hot Slugs should be stored in secure, remote location that can dissipate their heat safely and passively with low rate of physical deterioration, like either under a mountain or open-air in a desert. This should include the facility to re-glassify (re-mold) Hot Slugs shattered by wrecks or flaws. This could also perhaps be (part of) an original Hot Slug production facility. A railway branch may be needed to connect the remote storage facility to the network for work trains.

Power Converter assemblies require common manufacturing/maintenance skills and equipment. This is where all the moving parts are maintained and replaced. Since the assemblies are likely contaminated after long exposure to the Hot Slugs, this shop seems more likely to be located near an industrial town served by the railway network than in a major city. This is probably also the facility where the Hot Slugs and Power Converters are married and separated. The ideal location for such a heavy repair shop is near (but not too near) the yards serving a major terminal or junction that ISN'T a major city. Access to the supply chains of repair parts and heavy equipment, and to the disposal and recycling chains of used and contaminated parts are also vital, so don't stick this shop in a secret mountain bunker - it's a busy place and needs highway access.

The route for work trains between the Hot Slug storage range and the Power Converter Shop should avoid population centers and critical infrastructure. A circuitous-and-slow but safe route.

Note that having large radioactive power sources for locomotives means most populations won't want trains running through their town/city anymore - railway alignments may become circumferential with suburban terminals and yards. Shops and sidings may be spaced a bit farther from the mainline (and from each other) than historical practice for safety.

  • $\begingroup$ Similar considerations were being made when there was serious attempts to engineer nuclear powered aircraft as well. $\endgroup$
    – Thucydides
    Commented Jan 9, 2018 at 15:46
  • $\begingroup$ I am sorry, I do not see how this answers my question. Reading it again and again it is full of valid and good thoughts but it's not an answer. $\endgroup$
    – dot_Sp0T
    Commented Jan 9, 2018 at 18:32
  • 1
    $\begingroup$ It may not be the answer you expect, but it does address engineering complexity, recoverability, storage with minimal deterioration, and footprint: The Hot Slug must be separated from the Power Conversion assembly, and stored separately. I understand if you want to keep RHUs as a single, atomic unit...but then you are asking for a much more unnecessarily elaborate and expensive solution. $\endgroup$
    – user535733
    Commented Jan 9, 2018 at 19:47
  • $\begingroup$ @user535733 that's not even what I mean. It's that there's no actual proposed solution in the answer. It's just generalities which anyone spending a few minutes thinking will come up with. To put it in a metaphor: There is not much if any meat to the bone. I can't judge your answer by the criteria given, because I can't see them being addressed. This might be as much of my fault as yours, i just decided to communicate my side instead of sitting still. $\endgroup$
    – dot_Sp0T
    Commented Jan 9, 2018 at 20:00
  • $\begingroup$ Hmmm. Well, sorry if it's not what you are looking for. Sometimes answers to hard questions are like that. $\endgroup$
    – user535733
    Commented Jan 9, 2018 at 20:22

To do what you want, the first question is what radioactive material to use (probably not radium). The criteria are how efficiently it supplies heat, how much shielding it requires, and how long is its half-life. There is a very good discussion of possible fuels in the Wikipedia entry on radioisotope thermoelectric generators (RTGs). For your application, the best material is Americium 241. It requires only modest shielding and has a half-life of 432 years (so could meet your desire for RHUs that last centuries). Besides a longer half-life compared to Plutonium 238 (the standard fuel for RTGs), it decays emitting alpha and beta particles plus relatively low energy gamma rays and no neutrons. An RTG's power diminishes over time partly due to the half-life of the Pu 238 (87.7 years), but also because of the degradation of the thermopile material by the neutrons emitted from the Pu 238.

To generate mechanical power from the RHUs, you could use a Sterling cycle or other mechanical heat engine. They are reasonably efficient, but do have moving parts. For maximum longevity, you could use a thermopile (an array of thermocouples). At present, the best are only about 8% efficient, but there are no moving parts and, in the absence of neutron bombardment, should last many decades or possibly centuries. As with any heat engine, the cold sink is just as important as the heat source. In a low tech, recovering world, sticking the cold end into a fast running stream might be a practical solution.

Information on potential fuels for RTGs or RHUs:
https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_generator (see Section 3)

Data on Americium 241:

  • 1
    $\begingroup$ The material that is used is radium. As mentioned in this question and further described in the previous question. This question is not about how to generate power from the RHU, this question is about how to store an RHU as described in the linked question for decades up to centuries so they are still usable when recovered. $\endgroup$
    – dot_Sp0T
    Commented Jan 11, 2018 at 10:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .