Inspired by this question on putting massive capacitor-banks on spaceships I have been wondering about alternative ways to store excess energy from ever-running reactors.

In this question I would like to explore the possibility of synthesizing elements with the excess energy that could be then again used as an energy-source on demand.

Q: What fuel(s) could I synthesize with the excess energy from my reactors?

Answers are judged by the following criteria:

  • Stability: The more stable it is in regards to reactions or spontaneous explosions, the better
  • Longevity: A highly energetic element that decays in a few seconds is of no use, it must be possible to store it for longer amounts of time
  • Compactness: The higher the potential energy in relation to the element's density, the better
  • Efficiency: The less energy that is lost in the process of synthesizing and then burning the element again, the better

Setting: The question assumes a 20min into the future setting. Nothing mind-boggling or physics defying. The idea of synthesizing just seemed appropriate.

In regards to shutting down reactors instead: This is a valid point, and a good idea. Yet the goal of this question is to explore means of storing this excess energy that can amass between bouts of massive consumption. If it helps you, just think of these reactors as really complicated and hard to shut down1.

1E.g. a military vessel would not want to shut them down due to the complicated procedures and time involved in shutting them down and/or turning them on again.

  • 1
    $\begingroup$ As I understand it, you specifically want energy to be stored by chemical reactions, and other methods are out of scope. But it also seems that you are allowing nuclear transformations, or changes at the atomic level. Please clarify - chemical, nuclear, or both? And does it HAVE to be done in the nuclear reactor directly, or can it be a process that uses the energy from the reactor, although indirectly? $\endgroup$ Commented Nov 4, 2017 at 18:40
  • $\begingroup$ Is a process that is only 2% efficient preferable and more acceptable to a process that is only 1% efficient? Or is there a minimum efficiency that must be maintained? $\endgroup$ Commented Nov 4, 2017 at 18:50
  • $\begingroup$ @JustinThyme I have absolutely no idea what is possible, I couldn't find anything good on the subject beyond the idea of synthesizing something or similar instead of 'charging' things - thus the question. If the best possible has an efficiency of 2% then that is the best answer in that regard. $\endgroup$
    – dot_Sp0T
    Commented Nov 4, 2017 at 19:02
  • $\begingroup$ @dot_Sp0T Does it have to be synthesizing elements, or would molecules be acceptable? In places using a high percentage of renewable energy sources, excess electricity at certain times of day or year is already a problem today and has prompted the development of experimental technology to use that excess electricity to synthesize methane (~= natural gas), for which extensive transport and storage infrastructure already exists. See for example this 2013 article. Efficiency seems reasonable for your purposes. $\endgroup$
    – njuffa
    Commented Nov 4, 2017 at 22:52
  • 2
    $\begingroup$ I don't like shutting down reactors. The mortgage bill is exactly the same, fuel costs are negligible especially if you wise up and use thorium, and the maintenance difference is fairly small. Actually spool-up/spool-down is probably harder on maintenance than 24x7 running. $\endgroup$ Commented Nov 5, 2017 at 0:26

12 Answers 12


First thing coming to mind is $2 H_2O <-> H_2 + O_2$ electrolysis is rather efficient and so are fuel cells.

Probably by the time we get FTL such processes would be even better.

Other chemical processes can be used if and when suitably efficient reversible conversions are developed (possible candidates include: ethylene, hydrocarbons and Aluminum oxides). They may provide a better storage, erg/gr or other practical benefits.

I advise against electrical or kinetic storage because of high danger of immediate conversion to raw heat in case of any malfunction; OTOH chemicals can be stored safely in separate, jettisonable, containers outside the ship.

If this is not a concern then a large flywheel on frictionless magnetic bearings is the best solution for both conversion efficiency and power/weight.

  • $\begingroup$ I like this one because it is simple and endlessly reversible. $\endgroup$
    – akaioi
    Commented Nov 5, 2017 at 1:29
  • $\begingroup$ plus in an emergency the work as rocket fuel or biological needs. $\endgroup$
    – John
    Commented Nov 5, 2017 at 2:21
  • 5
    $\begingroup$ The trouble is storing the Hydrogen. It tends to exit most containers quite quickly. $\endgroup$
    – wizzwizz4
    Commented Nov 5, 2017 at 8:47

Consider power-to-gas technology. One variant of it, currently in the experimental stage, uses excess electricity to split water molecules by electrolysis, then combines the generated H2 with CO2 to create methane. Since natural gas is essentially methane with traces of other gases, the generated methane can be used in the same way natural gas is used. According to Wikipedia, the process has an overall efficiency of 30% - 40% depending on how the gas is ultimately used.

Obviously this technology can utilize the extensive transport and storage infrastructure for natural gas already in place in many countries today. This includes the use of compressed natural gas (CNG) as alternative fuel for automobiles.

  • $\begingroup$ @SherwoodBotsford I am not disagreeing and you may want to supply your own answer along the lines of your comment. I didn't perceive the question to be restricted to a spaceship environment, and tried to answer within the constraints of a "20 minutes into the future" setting, which implies (to me) the technology should be in the experimental stage right now. $\endgroup$
    – njuffa
    Commented Nov 6, 2017 at 16:56
  • $\begingroup$ Took your advice. $\endgroup$ Commented Nov 7, 2017 at 15:47
  • $\begingroup$ +1 - better than hydrogen answer because we know how to store methane, and still have huge problems with hydrogen. $\endgroup$
    – Mołot
    Commented Sep 24, 2018 at 15:13

Here is something we could do now. Use aluminum metal to store energy, and oxidize the metal back to the oxide to create power.

I proposed the concept here: http://www.halfbakery.com/idea/Thermite_20powered_20steam_20car#1379964483

1: Refine aluminum from plentiful ore using excess electrical power (or cheap power from hydroelectric sources). That is done.

2: Oxidize aluminum for power. The thermite reaction is a familiar example of the power stored in aluminum metal: in this reaction the aluminum strips oxygen from iron oxide to form the aluminum oxide, in the process emitting a lot of heat. Some people think that when this reaction really gets going it can strip the oxygen from water also.

3: GOTO 1: Refine aluminum oxide back into metal using plentiful electricity.

The earths crust has a lot of aluminum as oxides and ore. Refined aluminum metal has a tremendous energy density (the reason for recycling the metal, not refining new - it takes a lot of energy to make the metal). Aluminum is stable and does not blow up, as evidenced by empty soda cans and beach chairs. When you get it hot enough it oxidizes with a tremendous output of heat - the thermite reaction. Thermite uses iron oxide as the oxygen source but you could use forced air or maybe even water, once it got hot enough.

This would work better on the earths surface where there is abundant oxygen than it would in space, where you need to bring along the oxygen.


The Answer

Super-conducting materials with high electron density should (or would be after we refine the technology) meet most of your needs. Ta-Nb-Hf-Zr-Ti (Tantalium-Niobium-Hafnium-Zirconium-Titanium) based alloys are such materials and one such is [TaNb]1x(ZrHfTi)x. I'm not even going to pretend to know enough about the specific dynamics of these materials to comment further other than to point out that as superconductors they have (a) stability and (b) very high valence electron density.


Your question feels like a technology dichotomy. You have the technology to synthesize mass (basically reversing e=mc2, if I understand your question), but you don't have the technology to shut off reactors you're not using? We do that today, firing up natural gas, coal, even nuclear rod banks depending on electrical demand.

The creation of mass for energy storage would have substantial effects on the operability of the ship as its mass and center-of-gravity shift during the creation and utilization phases of these materials. Yes, you can (theoretically) design the ship to compensate... but why have the weakness at all when you can just shut unused reactors off?

(I understand from your question that the reactors are "ever-running," but that's the dichotomy.)

  • $\begingroup$ Rearranging e=mc^2, not reversing it. Solved for m, it becomes m=e/c^2. That is a lot of energy for not much mass. $\endgroup$ Commented Nov 4, 2017 at 18:46
  • $\begingroup$ @JustinThyme it is a lot of energy. The tricky bit is extracting that energy. If you can do that then you have an almost unlimited source of energy. $\endgroup$
    – Slarty
    Commented Nov 4, 2017 at 19:13
  • $\begingroup$ @JustinThyme and either way, it doesn't change my basic premise. If you have the ability to do that, why don't you have the ability to turn a reactor off? $\endgroup$
    – JBH
    Commented Nov 4, 2017 at 19:18
  • $\begingroup$ I am definitely not repudiating your point. The average person looks at e=mc^2 from the point that it is a lot of energy from a small amount of mass. They don't necessarily realize that the opposite is that it takes an awful lot of energy just to produce a small amount of mass. Most people think the reverse - that you can also get a large amount of mass from a small amount of energy. The 'reverse' doesn't work that way. But yes, shutting down the reactor would be much more prudent than using it to produce a very small amount of mass. It's more efficient to get more mass in the first place. $\endgroup$ Commented Nov 4, 2017 at 23:30

This doesn't answer you question exactly but may be useful idea.

IMHO any from of chemical energy storage is not going to be compact enough to power a FTL drive.

However you could store your excess energy as cool plasma in a couple of magnetic bottles (tokamak).

Using the energy from your power source create two, balanced, streams of plasma. One highly positively charged (protons from ionized hydrogen gas), the other negatively charged (mostly the electrons from the ionized hydrogen). You feed each stream of plasma into a different magnetic bottle (or alternating bottles). Using RF generators you 'cool' the plasma, so that it has less side to side velocity and just keeps moving around the loop. You could also also add more energy by increasing the velocity of the plasma around the loop.

This is very loosely what happens in the large hadron collider. Only the LHC accelerates a small quantity of plasma very very fast.

When It comes to igniting your FTL drive you open a magnetic valve and let the two streams of plasma at each other and zoom their excessive attraction for each other powers your drive and 'blip' you are in a new location.

The limitation to the amount of energy you can store is:

  1. Strength of the magnetic containment (solved problem from developing fusion generators)
  2. The stability of the RF cooling/acceleration system
  3. Bremsstrahlung radiation (X-rays from accelerating electrons (around a loop))

Point 1 & 2 are rather obvious, nuf said.

The Bremsstrahlung radiation will cause your magnetic capacitor, to leak/loose energy over time. This is why you would want to have lots of plasma moving slowly (rather than the LHC which has a little moving fast).

Further thoughts:

Rather than hydrogen gas you could use the helium 'ash' from your fusion generator in the plasma containment rings.

You could use small streams of this plasma to meet and power 'impulse' rockets, when you want to move about in real space.

You could use bursts of plasma as a short range weapon (the plasma will dissipate rapidly in space/ when not magnetically contained. ( this could be an unconventional/ last ditch tactic; as it would bleed the FTL charge you are building up)

You could cascade several containment units, so that the discharge from one is used to accelerate the plasma in the next, which is then used to accelerate the next etc etc. Kind of like double bouncing someone on a trampoline to then double bounce on a larger trampoline. The the loss from Bremsstrahlung radiation would not be a problem as you are not trying to store high speed plasma but cascade quickly for a FTL jump. The pattern of bursts of Bremsstrahlung radiation during a cascade would be a characteristic 'jump signature' of each (class of ?) vessel.

Cool buzz words warning:

Careful engineering of the plasma containment could let you use the storage system as a synchrotron laser (Yes that is a real thing) or the Bremsstrahlung radiation from one storage ring could be used as part of a klystron (yes that is also a real thing) to power the RF generator of the next stage of the FTL cascade.


eeStor had a prototype or scam (depending on who you talk to) that claimed to store 50 kWh in a box about the size of a bar fridge. It was done using barium titanate based capacitors. Energy densities were claimed at levels better than the best Li-ion batteries by weight and comparable by volume. Being capacitor based it could handle much higher charge/discharge rates than a battery, and was not subject to the short life of batteries (millions of cycles vs hundreds) It may be possible to do something really interesting with microfabrication.

Given materials with much stronger tensile strength than we have now, flywheel storage looks attractive. Potentially the energy goes up with the square of the tensile strength of the material used for the rotor.

Velkess attempted to crowd source funding for a flywheel system that featured low precision construction, which would markedly reduce the cost of production. Currently they need funding. http://www.velkess.com/flywheel.html

There is lots of merit in a methanol economy. Use surplus energy to make methanol; burn methanol in existing internal combustion engines. The modifications required are fairly trivial. Methanol is used in racing cars. The current methanol reaction is only about 60% efficient, giving an awful round trip efficiency. https://en.wikipedia.org/wiki/Methanol_economy

Splitting water into H2 and O2 then recombining in a fuel cell or or in a MHD turbine gives you better efficiency, but storage is bulkier, and it doesn't have the advantage of using present infra-structure.

There are various battery technologies that are too ponderous to use in vehicles, but are getting attention for utility scale power. Sodium sulfur, iron nickel are two. The latter is quite old, but has advantages in that it tolerates many more charge/discharge cycles.

Liquid electrolyte flow through batteries are getting attention right now. The reactants are stored as solutions, and react in a cell. The capacity of the battery is determined by the size of the tanks, the power determined by the size of the battery plates. Currently very expensive.


If we relax the 20 minutes in the future:

Silicon lithium promises a 10 fold increase in battery density, but is likely 10 to 20 years off. Li-Air and Aluminum air has similar potential density. Al-Air would be cheap, as Al is much more common than Li. http://www.visualcapitalist.com/future-battery-technology/

You are in space. Lots of room. How about very large superconducting coils. You are in essence storing power as a magnetic field. This can work at the surface too, but the magnetic fields may be a serious nuisance. (Crossing a magnetic field with a conductive object induces eddy currents in the conductor, making the conductor hot.)

Larry Niven in one of his stories talks about 'molecular distortion' batteries. Energy is stored by changing the shape of some stiff molecule. No other detail is given.

Robert Heinlein has 'Shipstones' It gets a mention in various places, but comes to the fore in "Friday" No tech is given.


There are lots of great answers here, about the specifics of how you can chemically store the excess energy. Let me throw in a few extra thoughts...

First, chemical storage of energy is a buffer, not a full solution. You only have so much aluminum or H2O or methane on board. What happens when it is all used up and your reactor is still running full-bore?

Perhaps you should consider other ways of limiting output.

Is your reactor such that you can throttle the fuel for it? This will help to stretch out the time it takes to "fill up" your chemical sink.

Does your design entertain the possibility of control rods? (Assuming a nuclear reactor here) They absorb some of the wayward neutrons, damping down the reaction. Your normal operation might have control rods halfway extruded into the reaction chamber, allowing a moderate reaction & energy flow. During peak conditions you can withdraw the rods, getting you max flow. In an emergency, drop the rods all the way in, quelling the reaction altogether. You need a mechanism like this anyway, for disaster avoidance; why not use it to mitigate your other problem, too?

Worst-case scenario...

All your spare H2O is split. Control rods don't work in your reactor. You can't starve the reactor of fuel. What can you do?

Well, maybe you can ... work with that energy. Have banks of heavy millstones; spin them to use the excess power. They'll slow down on their own, so you can spin 'em again. Basic idea here: you need to avoid your longevity requirement as an ultimate backup, and store energy in a way that won't just pile up.

Finally, if you still are generating too much electricity, use it to power as many refrigerator lasers as you need to get rid of all that energy. Saw a discussion where they seemed to think it'd work ( https://www.physicsforums.com/threads/is-a-refrigeration-laser-thermodynamically-possible.313229/ ). Um. Don't do this when you're docked to a space station.


According to this page, compressed air energy storage (CAES) looks safest, most durable, and most useful (emergency air supply, thrust):

Comparison of energy storage technologies

Not asked for, but you could use a loading bay or double hull to recycle the air, both would enable leak detection before recompression.

  • $\begingroup$ On the other hand, it takes a LOT of space to have enough uncompressed air for this system to work in. And a lot of space equals high area for micrometeorites to strike and make a hole leaving this system not working at all. $\endgroup$
    – Mołot
    Commented Nov 6, 2017 at 17:46
  • $\begingroup$ @Mołot That's rather presumptuous. Ice shielding of the loading bay can help, and damage can be repaired. $\endgroup$ Commented Nov 7, 2017 at 0:26
  • $\begingroup$ Please, edit your answer to provide additional information. Comments are "second class" citizens here and may be purged without notice, explanation and chances for undeletion. By the way, damage can be repaired but you can't suck escaped air back. $\endgroup$
    – Mołot
    Commented Nov 7, 2017 at 6:33

No one seems to have mentioned antimatter synthesis yet. Antimatter wins hands down for highest energy density and is also stable, though admittedly it's tricky to store and prone to spontaneous detonation should its storage system fail. . . .

Anyway, it's possible to produce small amounts of antimatter today (with rather terrible efficiency currently), so it's an option if you really have nothing better to do with the energy from your always-running reactors. And, as a bonus, it would make an extremely high-power energy source for an FTL drive.


If you really into nuclear synthesis, you should consider what breeder reactors do.

Most of the historical effort for breeding has been using neutron capture to transmute U-239 into Pu-239 (after a few intermediate decays). Specifically U-238 + n -> U-239 -> Np-239 -> Pu-239. Pu-239 is fissile whereas U-238 is fertile -- you now have useful fuel for reactors (or bombs).

Th-232 is also fertile, but after capturing a neutron eventually becomes fissile U-233, another useful reactor fuel.

Neutron capture can occur directly via exposure to an active nuclear reactor, and it id quite difficult to accelerate neutrons since they are neutral particles. So, you may quibble that this is not a result of using excess energy production of a nuclear reactor.

Fee protons are easily accelerated, but they are difficult for use in growing larger nuclei, since they are strongly repelled by the nucleus.

If you want to consider practical nuclear synthesis, these 2 reactions are really the only practical options you have. Either reaction could be made to work on a fairly large scale, though Th-232 is considerable easier because it has a relatively large neutron cross section in the thermal region.

You don't really need or want to build large stockpiles of the converted elements though, as you are able to synthesize the fissile materials as needed (with enough extra in the pipeline for processing the generated fuel). A liquid fuel reactor is desirable because you can extract the desired product on a continuous basis, instead of allowing it to absorb additional neutrons (you don't want Pu-240 or U-234).

Many countries have experimented with breeder reactors (mostly to generate Pu-239) and have eventually given up on them. The Thorium cycle is actually more practical from a physics and engineering viewpoint, but has not been historically popular for breeding, though there is renewed interest in the thorium fuel cycle in recent years.


Chemical reactions fail in the compactness department. The energy output of the nuclear reactor is orders of magnitude beyond any chemucal reaction.

You might consider my answer to What's strongest non-nuclear explosive I can make with nanotechnology? and My question on developing this near-future technology. Both posts have the same xkcd cartoon illustrating my statement in the first paragraph.

178m2Hf for example stores 2.446 MeV per atom, compared to over 200 for Uranium fission. Adding a bar for 1 330 000 MJ/kg to the above chart (the bar will be 260 m tall), we see that even at 1% of nuclear levels it's about 10,000× more energetic than any chemical energy.

The “best” storage, which is equivalent to slowing the reactor minus efficiency losses, is to spend the energy to reverse the reaction used for power. E.g. split He back into Hydrogen. If you saved the He waste, you reverse that to return to your fuel store.


The normal solution to this question is handled from an economic viewpoint rather than an engineering one.

The reasons why are fairly simple to see. Consider a few facts, then an economic picture emerges:

  1. Economies of scale usually make a single large power plant cheaper per unit energy produced than a large number of small power plants.
  2. The cost per unit power tends to rise unless the plant is operated at a significant percentage of its capacity.
    You still incur the same debt service and maintenance operations whether you run at 50% capacity or 80%, so 80% means more revenue for the same costs.
  3. The laws of thermodynamics are still going to apply to any storage technology employed---so you have efficiency loss in storing and efficiency loss in recovering stored energy.

So the big picture is to build power plants that are scaled to operate at 70-80% capacity in a quantity that provides for running peak demand by running close to capacity. Or by running four plants close to peak and periodically bringing online a peak demand power plant, also running at close to capacity.

Granted a spaceship is a use case that is entirely different from the demands of generating and distributing power on a large scale to a population. There the engineering problems must account for making a worst case demand available continuously, so best efficiencies of dollar per watt/hour fly out the window. For that matter, the wisdom of adding a lot of expense and extra mass (which requires bigger thrust to move the ship) flies out the window as well. The direct and indirect costs of a storage mechanism for a low exirgy source of energy outweighs the benefit.

Money and physics can say a whole lot about what will or won't ever be attempted, even if it is possible.

The steverino paraphrase of the three laws is thus:

  1. There is a game.
  2. You can't win.
  3. You can't break even.
  • $\begingroup$ Just build your heavy things in space or on the moon. $\endgroup$ Commented Nov 5, 2017 at 23:49
  • $\begingroup$ Why the downvote? The OP clearly stated that current tech was required. "20min in the future" That makes my point that there is no economically viable alternative a sound answer. If there was already such an alternative it would be heavily used in preference to building oil,coal, natural gas or nuclear peak demand generators. It looks to me like I got a downvote by someone who doesn't even understand the answer. $\endgroup$
    – steverino
    Commented Nov 6, 2017 at 0:05
  • $\begingroup$ Sorry, I thought it was about spaceships. Anyway, there are plenty of energy storage systems in use today. $\endgroup$ Commented Nov 6, 2017 at 0:14
  • 1
    $\begingroup$ Let us continue this discussion in chat. $\endgroup$
    – steverino
    Commented Nov 6, 2017 at 0:45
  • 2
    $\begingroup$ This answer is being discussed on Worldbuilding Meta in help me understand downvote. $\endgroup$
    – user
    Commented Nov 6, 2017 at 8:49

You must log in to answer this question.

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