Can a nuclear rocket engine create electricity?

[Minmus Taster]

Could a reactor in a nuclear engine also be used to create electricity for the spacecraft using it while also serving its intended purpose?

[magnemoe]

I say yes but it had to be designed for thins. I guess you could set power to low and close the chamber and then run it gas cooled. 
So you run gas trough it but you then run it trough an turbine and cooler for reuse. 
It would be lighter than an separate reactor for power. 

[tg626]

The reactor provides heat for propulsion, heating the hydrogeny fuel. Reactor heat isn't something you just turn on and off, so its going to be sitting there radiating heat that would otherwise be waste, so you could harness that to do other things - at least when you aren't using it for propulsion - but then the question would become if it's worth its weight to do it vs other methods which could be presumably used at all times.

[kerbiloid]

Propulsion mode: heat the fluid and throw it out to produce thrust, use a direct heat-electricity exchanger as a low-power source of electricity.

Energetics mode: heat the fluid and run t in a turbine loop, generating high power.

Of course, you should design the engine system this way.

[DDE]

11 hours ago, Minmus Taster said:

Could a reactor in a nuclear engine also be used to create electricity for the spacecraft using it while also serving its intended purpose?

Yes, full stop. Designs for solid-core NTRs with an integrated Brayton-cycle generator have been considered at the level of NASA DRMs. Some of those designs also include the lOx "afterburner" from bimodal NTRs to produce a trimodal NTRs.

Higher-temperature designs like GCNTRs can feature an MHD coil in the nozzle to syohon some of the exhaust's energy as electricity. In fact, commonality with power generators seems to increase as you go into higher ISP - the Energomash RD-600 was supposed to share its core with the Energia EU-610 gas-core reactor, while most fission fragment rocket designs are essentially dusty plasma reactors with one end unplugged.

[K^2]

It should be noted that you absolutely have to take an efficiency loss somewhere to generate electricity from an NTR, but for a lot of the missions, the efficiency loss might be insignificant compared to the advantage of not having to rely on some other source of electric power.

You do have a bit of a choice of what sort of efficiency loss to take. If you try to utilize heat flow from the fuel to heat exchanger, which is trivial enough with some NTR types, you will have to take a temperature drop in the exchanger, resulting in slightly lower I_SP. If instead, you divert some of the reactor heat, you can keep the exchanger just as hot, but you'll need a slightly larger reactor, making the engine heavier, taking the cut in TWR. Finally, if you go for an MHD generator, you'll be slowing down exhaust, which will harm both the TWR and I_SP.

[tomf]

Many missions with nuclear engines are going to involve long coast phases so you probably want a design where you can produce electricity without thrust. The ones that tap the exhaust aren't going to be able to do that.

[RCgothic]

18 hours ago, tg626 said:

Reactor heat isn't something you just turn on and off

Fission can just be turned on and off.

~6% operating thermal power comes from fission product decay though, reducing to about 0.1% over ~40 days, so.

You don't necessarily have to provide cooling for decay heat if you don't care about the state of the engine after it's finished firing (propellant exhausted, non-earth intersecting stage trajectory, doesn't matter if it melts.) Fission fragment engines avoid this altogether by exhausting the fuel, so the engine isn't going to get hot from spent fuel emitting decay heat inside the engine 

But there's no reason not to tap some heat off the power cycle (or decay heat cooling) for electricity generation if you don't mind bringing a little extra mass along. 

Edited March 25, 2023 by RCgothic

[tg626]

2 hours ago, RCgothic said:

over ~40 days, so.

Which is why i said you cant just turn it off and on, by which I meant like a normal LFO engine or light bulb. It's ON / It's OFF.

Using the light bulb analogy, it's a light bulb that takes 40 days to slowly dim to dark.

This predicated on the reader not necessarily understanding the finer points of nuclear fission.

One of which being neutron poisoning of the fuel mass if you DO attempt to just halt the reaction immediately (SCRAM the reactor) and then you still have decay heat to contend with.

[sevenperforce]

My standard concept for any sort of near-future spaceship propulsion is a trimodal nuclear-thermal NTR running on something like methane or propane (for reduced tank volume) with a LOX supply for afterburning and a closed-loop methane radiator system for generating electricity from the reactor during coast.

[K^2]

9 hours ago, sevenperforce said:

with a LOX supply for afterburning

That clearly works from the energy perspective, but I instantly have a lot of "How?" questions. Like, how do you inject oxygen into that (also run it through NTR to get the gas up to speed?), how do you design an afterburner that works at these temperatures and flow speeds, and how do you build a nozzle that survives this kind of an abuse? Do you happen to have a specific design in mind, perhaps, that you can point to? None of these problems seem unsolvable, just really hard, especially in concert, and if somebody already put in the work to try and figure it out, I'd be interested to see what they came up with.

[AckSed]

Aerojet Rocketdyne derived the Thrust-Augmented Nozzle (essentially an afterburner for a rocket engine) from their work on LANTR (LOX-Augmented Nuclear-Thermal Rocket) and its LOX injectors. It was up to test-stand tests last thing I read.

[sevenperforce]

29 minutes ago, K^2 said:

That clearly works from the energy perspective, but I instantly have a lot of "How?" questions. Like, how do you inject oxygen into that (also run it through NTR to get the gas up to speed?), how do you design an afterburner that works at these temperatures and flow speeds, and how do you build a nozzle that survives this kind of an abuse? Do you happen to have a specific design in mind, perhaps, that you can point to? None of these problems seem unsolvable, just really hard, especially in concert, and if somebody already put in the work to try and figure it out, I'd be interested to see what they came up with.

The idea has been kicked around for a while, but I believe one of the seminal treatments of the concept came from NASA in conjunction with Shuttle companion proposals:

https://ntrs.nasa.gov/api/citations/19950005290/downloads/19950005290.pdf

I'm not sure how far they got into specific designs, but it did include some pretty decent schematics:

The report talks at length about how they proposed to solve the supersonic combustion issues.

 

[kerbiloid]

The temperature of several kK leads to ionization of atoms...

The ionization breaks moleculs.

The energy required for the molecular assembly and disassembly is equal...

[sevenperforce]

11 hours ago, kerbiloid said:

The temperature of several kK leads to ionization of atoms...

The ionization breaks moleculs.

The energy required for the molecular assembly and disassembly is equal...

That's why another term for afterburning is "reheat".

As high-pressure, high-temperature exhaust gases pass through the throat of a de laval nozzle, they expand and accelerate. That expansion causes the gases to cool. In the case of a NTR, you're dealing with a flow of hot hydrogen, and so injecting liquid oxygen inside the nozzle will result in combustion at the boundary layer which reheats the gases and gives them more opportunity for expansion and acceleration out the nozzle.

That's why you can't inject the LOX inside the combustion chamber.

[kerbiloid]

23 minutes ago, sevenperforce said:

As high-pressure, high-temperature exhaust gases pass through the throat of a de laval nozzle, they expand and accelerate. That expansion causes the gases to cool. In the case of a NTR, you're dealing with a flow of hot hydrogen, and so injecting liquid oxygen inside the nozzle will result in combustion at the boundary layer which reheats the gases and gives them more opportunity for expansion and acceleration out the nozzle.

And you have to spend 8 kg of O2 per 1 kg of H2, raising the molecular mass from 1 or 2 up to 18, thus: dramatically decreasing the ISP and increasing by an order of magnitude the spent mass, in the rocket equation.

So, it eats the ISP of pure hydrogen. So in the best case it can replace high ISP with high thrust for the same engine, but you don't need it for NERVAs.

If the engine is really hot and the exhaust speed is tens km/s, it's just a bombardment of an oxygen target with hydrogen ions, The oxygen will be just ionized, and turn into a hot cloud inside the nozzle, pressing it from inside.

[sevenperforce]

22 hours ago, kerbiloid said:

And you have to spend 8 kg of O2 per 1 kg of H2

Or you can just use less LOX.

No hydrolox engine runs stoichiometric anyway.

[kerbiloid]

9 minutes ago, sevenperforce said:

Or you can just use less LOX.

Less LOX gives less energy, but still increases the molecular mass and thus decreases ISP.

[sevenperforce]

1 minute ago, kerbiloid said:

Less LOX gives less energy, but still increases the molecular mass and thus decreases ISP.

That's the point of afterburning -- temporarily increased thrust at the cost of temporarily decreased efficiency.

[kerbiloid]

26 minutes ago, sevenperforce said:

That's the point of afterburning -- temporarily increased thrust at the cost of temporarily decreased efficiency.

And the construction complexity and mass in case of the low-ISP (because the exhaust temperature should be below the ionization and dissociation of molecules) nuclear engine, additionally equipped with afterburner and oxygen propellant system.

In case of the high-temperature high-ISP engine, the molecules won't even appear. 

[wumpus]

On 3/30/2023 at 11:05 AM, sevenperforce said:

That's the point of afterburning -- temporarily increased thrust at the cost of temporarily decreased efficiency.

And presumably burning through all your LOX during ascent to orbit so you have lower total mass when you want higher efficiency (and don't need a separate high-thrust second stage).

Also, to answer the original question about "creating your own electricity", consider the link:

https://news.mit.edu/2022/thermal-heat-engine-0413

This is wildly better than the traditional method of thermocouples (although I'm sure NASA has considered the "old fashioned non-wide band gap" cells).

The article also suggests using such high-efficiency cells in conjunction with solar concentrators (which would absorb the entire spectrum into heat).  Assuming land usage isn't an issue, I'd suggest forgetting about maximizing efficiency and simply using a single layer for low cost (perhaps overdoing it, in this case the cost of the non-solar parts might drive the cost enough to warrant multiple layers, if not maximum efficiency) and building the concentrator large enough to run the cells 24/7.  I'm probably missing the costs involved in absorbing and storing solar heat at 2000K-3000K.  You might need high efficiency if you are getting your heat from natural gas, but with low cost mirrors (silver plated aluminum is relatively cheap) it comes down to the area available and just how hot you need to make your TPV work.

[Nothalogh]

On 3/25/2023 at 6:30 AM, tomf said:

Many missions with nuclear engines are going to involve long coast phases so you probably want a design where you can produce electricity without thrust.

That means you just aren't producing enough thrust to go with all that ISP.