Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Nitrous oxide energy storage

  1. Jun 16, 2016 #1

    I took part in discussions on various forums on viability of large-scale solar power as a primary energy source (on Earth), and on in-situ resource utilization (ISRU) for space missions. One idea discussed for Mars base was to produce CO and LOX from Mars CO2 atmosphere, obviating the need to find, scoop and process soil/ice/etc.

    I thought "hmm, interesting that Mars has this possibility to make fuel simply from air, while we on Earth do not".

    Then I realized that this is actually not true, there are compounds which can be synthesized from N2 and O2 which then can be used as fuel. Availability of some water vapor open up more possibilities.

    We don't use these fuels today on large scale because they are not very convenient, we have better ones derived from oil and gas.

    However, this may change. Large-scale solar power is finally happening. PV is no longer a "fancy green toy power source, not really competitive". It is competitive, and improving still. And as its market share grows, it will encounter the problem of needing to store energy during daytime for night-time load.

    So, to the question.

    Let's say we have huge PV installations whose peak power is significantly more than demand, making peak daytime electricity basically free.
    Let's say these installations are in Sahara and they are powering Europe.
    Can this excess power be used to produce fuels from the air, and use them at night to produce power? What those fuels can be? Is it economically viable?

    One possible fuel is N2O, nitrous oxide. It is non-toxic, it is not too difficult to store (and we already practice storage of many ton tanks of it), it is a monopropellant with a completely harmless exhaust. Questions: do current production processes for N2O use only air/N2/O2, or they use other routes?

    Another possible "fuel" is simply compressed air. I know that compressed air energy storage is practiced, but it has a problem: energy is lost when heated compressed air loses heat while in storage, and then on decompression it becomes very cold, nearly to liquefaction temps - which severely impacts amount of extracted energy. Because of this, commercialized compressed air energy storage schemes add a bit of natural gas burning to the generating machinery. This makes it "not exactly energy storage": now you need other fuel.

    Well, we have that N2O turbine running, don't we? And its exhaust is *warm* N2/O2 mixture. Can we use this heat to fix "decompressed air is cold" problem and use both schemes?

    Next. Compressing air allows to collect ambient water vapor. It can be used to produce other compounds: ammonia NH3, hydrazine N2H4, hydrogen peroxide H2O2. Can those be used as a fuel? H2O2 and N2H4 each can be used as monopropellant, and variously combined with N2O - as fuel/oxidizer pair. Since water is less abundant in air than N2/O2, and variable, I would imagine some sort of "N2O decomposition augmented by N2H4 burning" may end up being the best scheme.

    Other nitrogen oxides can be synthesized too. N2H4 + N2O4 is actually a quite potent fuel/oxidizer pair. There are toxicity and corrosion problems, however.

    Can you guys comment on economics and synthesis-from-only-O2-N2 difficulties for these compounds?
    Last edited: Jun 16, 2016
  2. jcsd
  3. Jun 16, 2016 #2
    P.S. I do know that N2O and H2O2 are not, strictly speaking, "fuels" - they are oxidizers. Here I use the word "fuel" somewhat loosely.
  4. Jun 16, 2016 #3


    User Avatar
    Science Advisor
    Gold Member

    Yes. This is what plants do. It's called photosynthesis. You take CO2 and H2O from the air and use solar energy to turn them into sugars. The sugars are high energy fuels which the plants can then oxidize via respiration and oxygen from the air to release the stored energy. In the non-biological world, there are methods to make fuels from CO2 in the atmosphere: https://en.wikipedia.org/wiki/Artificial_photosynthesis These methods generally focus on making fuels like methanol/ethanol, which are simpler than sugars and can used to generate power directly in fuel cells.

    While storing excess energy generated from solar power is still a major engineering challenge, there are many ways to go that are far more efficient than this. First, you mention compressed air: compressing the air is itself a way of storing energy: https://en.wikipedia.org/wiki/Compressed_air_energy_storage In addition, there are all sorts of other ways to store energy: batteries, thermal storage (molten salt), pumped hydro (you use excess energy to pump water up a hill, which stores it as gravitational potential energy). Most of these methods are 1) scalable, and 2) quite cheap.

    Oxygen is a pretty good oxidizer, and our atmosphere is 21% oxygen. For industrial/economy-of-scale purposes, it makes no sense to create an exotic oxidizer from air when the air itself will work just as well.
  5. Jun 16, 2016 #4
    CO2 concentration in air is very low. In my hypothetical example of vast PV power plants in Sahara, a way to store energy is needed which does not rely on water (so pumped storage is out) or industrial sources of CO2 such as coal-fired plants (the whole idea of PV is to stop using non-renewables).

    I mentioned that I know about it. Compressed air storage doesn't seem to be very successful.

    Batteries are not (yet?) cheap enough.
    Pumped hydro requires favorable geography and water resources, with requirements not exactly matching those preferred by PV. My example of "huge PV installations in Sahara" certainly can't use it.

    I don't propose to synthesize N2O because it's an oxidizer. I propose it because it is a *monopropellant* and can be decomposed with energy release. O2 is not / can not. Please try to read my post with more attention to my logic.
  6. Jun 16, 2016 #5


    User Avatar
    Science Advisor
    Gold Member

    Sure, why not.
    I'm not aware of any process that produces N2O from air. They almost always come from other sources. Maybe you're thinking of the Birkeland-Eyde process, which produces NO: https://en.wikipedia.org/wiki/Birkeland–Eyde_process There might be a closed-loop process where you can take ammonia from Haber-Bosch and use it to produce N2O on an industrial scale, but this still requires synthesizing ammonia (presumably from N2 and possibly-fossil-fuel-derived H2).
    Gas can be expanded isothermally (at least 1 startup company, SustainX, is looking into isothermal compressed air energy storage). But you're right; it'll never be 100% efficient. But then again, neither will anything else. In addition, this expansion cooling (where the Joule-Thomson coefficient is positive) is more pronounced for less-than-ideal gases (such as air). For gases closer to ideality (hydrogen and helium), this is less of a problem. Don't forget, too, that compressing the gas causes it to heat up (since compression is the opposite of expansion), and that if you store this heat, you can use part of it to rewarm the gas as it cools on expansion (but again, not 100% efficient).
    This might obviate the need to generate heat during expansion, but you'll have to use more than that amount of energy to synthesize the N2O in the first place. In addition, the N2O → N2+O2 reaction isn't barrierless at room temperature: you have to add heat to get this reaction to go in the first place. Finally, the exhaust might be warm immediately after the reaction, but if you let the products freely expand, they'll cool just as much as air will, because they are air. I have no idea on balance whether this will be more energy-efficient than current compressed air storage, but I doubt the advantage will be enough for it to be economical.
    So are you compressing air and then reacting it to make N2O? As I indicated before, I'm not sure the direct reaction is possible. Again, maybe a closed cycle of
    but I'm not sure what the efficiencies/energetics are for all of these reactions.
    Back of the envelope, CO2 concentrations on Earth (~0.0004% @ 1atm) are roughly 1/10 of the CO2 concentrations on Mars (~95% @ 0.006 atm). I suppose you could make the argument that the concentration on Earth is too low for making fuel whereas the concentration on Mars is just fine, but I think it's a number worth pointing out.
    While geography certainly can make pumped hydro easier or harder, it's not the final determinant. Water towers are pumped storage as well. But, let's say you want to restrict this discussion to the hypothetical "middle of the Sahara" (and nowhere near any mountains, and ignoring the fact that most of the Sahara is in fact rocky and at an elevation of ~1km above sea level); then in that case, thermal is probably the best bet for energy storage.
    ...yet. I'm not sure that should be considered a mark against them. I'm not sure what the funding structure for energy storage looks like, but it's probably heavily skewed toward batteries. On top of that, grid-scale energy storage is a relatively new problem in the grand scheme of things, and if someone had solved it by now, you probably wouldn't have started this thread.

    Overall, my gut says that a scheme where you compress air and react it to form N2O for excess energy storage, then decompose the N2O and decompress the air to release energy seems like it adds a whole bunch of potentially difficult steps to an already existing process for a questionable increase in efficiency.
  7. Jun 16, 2016 #6
    Yes, something like this. Since I'm not familiar with industrial-scale chemistry, I came here to ask "what would it take to make N2O from air? How much that would cost?" etc.

    Correct. However, on Mars, if you want to get CO2 from the air, you already have 95% pure CO2, whereas on Earth you need to remove 99.9996% of "contaminants" ;)

    Not a problem: "In the presence of a heated catalyst, N2O will decompose exothermically into nitrogen and oxygen, at a temperature of approximately 577 °C. Because of the large heat release, the catalytic action rapidly becomes secondary as thermal autodecomposition becomes dominant."

    Yes, that's the idea.

    Consider that on this scale compressed air storage needs either huge, huge tanks (in reality, huge underground caverns are planned for use), or high pressures: to approach density of a liquid, you'd need an impractical pressure of ~800 atm.
    Whereas N2O is stored as liquid at moderate refrigeration and pressure, making it much denser than compressed air.
    IOW: N2O has much better energy density.
  8. Jun 17, 2016 #7


    User Avatar
    Science Advisor
    Gold Member

    Energy-wise, it's whatever the heat of formation of N2O is from the elements. NIST is telling me it's about 82 kJ/mol. Scaling up a multistep synthesis of N2O and trying to keep the whole thing from blowing up/leaking everywhere would probably cost quite a bit.
    This is true (you're also helped out by the lower temperature), but it's not terribly difficult to distill CO2 from large quantities air (especially if you're freely expanding, and therefore Joule cooling, the air).
    1) If you're (say) in the middle of the Sahara, is energy density really an issue?
    2) If it's chemical energy storage you're after, why not just use the excess energy to e.g., electrolyze water (thereby storing the energy as hydrogen gas)? It's something we already know how to do pretty well at large scale.
  9. Jun 17, 2016 #8
    Energy density is an issue, yes. In this hypothetical future where solar power become the cheapest, and therefore increasingly dominating energy source, we have a new problem: we need to ride through the night. This means generating many gigawatts of power for many hours. How big the compressed air storage tank needs to be to power a 1 GWe generator for, say, 12 hours?

    Electrolyzing water needs water, while the best PV sites are in deserts. The sunnier the better. Which usually means "very dry places". Electrolyzing water produces hydrogen, which is a gas, thus having the same problem of having low density, thus needing huge tanks.
  10. Jun 17, 2016 #9

    http://www.powersouth.com/files/CAES%20Brochure%20 [Broken][FINAL].pdf


    Compressed air storage system with natural gas burning assist.
    110 MWe generator.
    19.8 million cubic feet storage cavern = 560673 m^3 ~= box 80x80x80m
    sized for 26 hours of operation.
    Pressure 1100 pounds per square inch ~= 75 atm.
    Last edited by a moderator: May 8, 2017
  11. Jun 17, 2016 #10


    User Avatar
    Science Advisor
    Gold Member

    The nice thing about water is that it's cheap and liquid, so you can store it in underground tanks. If you were to use hydrogen as an energy storage system, you could easily have a dedicated tank of water that you electrolyzed to hydrogen, without having to worry about being near an actual body of water. In addition, the average humidity in the Sahara is 25% (believe it or not).
    EDIT: The other nice thing about water is that if your tank leaks, it doesn't matter. The same doesn't apply for N2O, a relatively stable, incredibly potent greenhouse gas.
    This is probably easier to do with numbers. Assuming that NIST is right, the max energy you can get decomposing N2O into the elements is 82 kJ/mol (this assumes a barrierless reaction). And the molar mass of N2O is 44 g/mol. Which gives a specific energy of ~1.9 kJ/g, about the same as a non-rechargeable lithium battery. Compare that to 142 kJ/g for compressed hydrogen. In terms of energy density, liquid N2O has roughly the same density as water (it actually turns out to be ~1.2 g/mL, cf. water @ 1 g/mL), so the energy density is roughly 2.3 kJ/mL. For a lithium battery, the energy density is around 4.3 kJ/mL, and for compressed hydrogen, the energy density ends up being 5.6 kJ/mL according to Wikipedia. So around the same order of magnitude. What I can't tell from what I've read is if the figure for hydrogen includes only the chemical energy for H2 combustion, or if it also includes the energy of expansion of the compressed hydrogen. If it's only chemical energy, compressed hydrogen could be significantly more energy dense in practice than the number I quoted here. If not, then it's roughly comparable to the number I calculated for N2O.

    Side note: If you're allowed to make liquid fuel from carbon, then you get much higher energy densities (e.g., gasoline @ ~34 kJ/mL).
  12. Jun 17, 2016 #11


    User Avatar
    Science Advisor
    Gold Member

    One more thing: if you're using the cycle that I mentioned before to make N2O, you're already making H2 as an intermediate on the way to ammonia. Like I said before, I'm not sure if you can make N2O directly from the elements. As far as I know, there is no industrial scale way to do so right now. But the broader point of energy density is a good one. If you don't gain any energy density above already-existing processes even in the best-case theoretical scenario, what other advantages would a route through N2O give that existing energy storage methods would not? and would that be enough to outweigh the drawbacks? I doubt it.

    The thinking is good, though: storing excess energy in energy-dense chemicals derived from ubiquitous sources. Which is why I keep harping on CO2 to methanol or something similar. An approach like this is probably best suited for transportation or individual-scale energy storage applications. For something like grid-scale storage in the middle of a desert, I'm not so sure. But then again, I still think arrays of water towers (as pumped hydro storage) might be a decent, environmentally friendly solution to explore (my fellow chemists would be ashamed).
  13. Jun 17, 2016 #12
    CO2 to methanol needs extracting CO2 _and_ a source of H2 (probably water).
  14. Jun 17, 2016 #13
    Or 2.3 MJ/L, or 2.3 GJ/m^3.

    To produce 1 GW for 24 hours, ignoring all losses, you need 86400 GJ of stored energy. With energy stored in N2O at 2.3 GJ/m^3, that's 37560 cubic meters of storage, or 332 railroad DOT-111 tank cars.

    "Only" about 170 railroad DOT-111 tank cars filled with lithium batteries? :) That might be a tad expensive.
  15. Jun 17, 2016 #14


    User Avatar
    Science Advisor
    Gold Member

    As I mentioned before, N2O production also needs a source of hydrogen (probably water). In addition, who says you need to extract CO2? Why not just store the CO2 and H2O that you need at the point of methanol generation? It's very easy to store both carbon dioxide and water. Then convert them to methanol/ethanol/whatever using excess energy from your solar array. You wouldn't have to worry about nitrous oxide at all. With energy stored as methanol (I think the energy density was ~15 GJ/m3, to use your units), you'd only need about 50 of those train cars. 1 mole of methanol combusts to 1 mole of carbon dioxide and 2 moles of water, which all have roughly the same density (around 1 g/mL ± 0.5 g/mL or so). So if you want to store all the methanol, carbon dioxide, and water separately, you'd need ~200 of the train cars. And you'd avoid all the safety concerns surrounding nitrous oxide.
  16. Jun 17, 2016 #15
    N2O molecule has no hydrogen. I hope it's possible to have a closed cycle plant with only O2 and N2 as input.

    Wait a sec. Where are you going to get CO2 in Sahara?
  17. Jun 17, 2016 #16
    Ant there is no way to get the energy out of the desert?
  18. Jun 17, 2016 #17


    User Avatar
    Science Advisor
    Gold Member

    Even if it is possible, it's still not as energy dense as compressed hydrogen.
    The same way we get solar panels into the Sahara: by carting it out there (possibly on 200 train cars).
  19. Jun 17, 2016 #18
    Solar panels are installed once and work for decades.
    For 1 GWe generation during night, you'd need to process hundreds of train cars of CO2 *every day*.
  20. Jun 17, 2016 #19
    This might work. Something like electrolysis plants on the Mediterranean/Atlantic seashores...
  21. Jun 17, 2016 #20


    User Avatar
    Science Advisor
    Gold Member

    No, you wouldn't. That was my point. You truck carbon dioxide and water out there once and then you cycle between methanol and CO2/H2O to store and release energy.
    EDIT: Just to be clear: you use the CO2/H2O to make methanol using extra solar energy. Then you combust the methanol at night to make CO2/H2O. The system is closed (with the exception of adding/subtracting O2).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted