Improving space nuclear reactors

[RobertGC]

I am interested in nuclear reactors for space propulsion. I have a few questions about it. Nuclear fuel cells in ground based nuclear power plants can work for years before burning out. Why is it for nuclear space reactors they only last for hours?

Also how can we improve the proportion of the nuclear fuel that actually gets consumed before we have to dispose of the fuel cell? I understand the fission reaction gets blocked by the fission products which inhibits the reaction from continuing. Couldn't we just have the reaction proceed in layers for a fuel from the outside in? When the outside fuel got consumed we would remove that and allow the inner layers to react. If we make the layers thin enough we should get little blockage from the fission-products.

If we could get most of the fuel to react then we would need much less radioactive material to be sent to space so nuclear space propulsion would be much less controversial. Also if the ground based reactors could use most of the fuel, much less uranium would be needed to operate a reactor and you have lower operation costs and radioactive waste being produced.

Bob Clark

 

[SteamKing]

I am interested in nuclear reactors for space propulsion. I have a few questions about it. Nuclear fuel cells in ground based nuclear power plants can work for years before burning out. Why is it for nuclear space reactors they only last for hours?

I'm not sure what a "nuclear fuel cell" is.

Nuclear reactors on Earth use nuclear fuel pellets encased in metal tubes. Bundles of these tubes are then inserted into the reactor core, where the nuclear fission reactions release heat.

https://en.wikipedia.org/wiki/Nuclear_fuel

Satellite reactors do not operate on the same scale of power generation as a nuclear plant on earth. Most of these reactors are known as radioisotope thermal generators, or RTG for short. They are capable of producing a few hundred watts of power when new, which power declines gradually as the radioisotope decays:

https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_generator

The RTGs aboard the Voyager spacecraft operated for almost 25 years after launch.

Most satellites in Earth orbit use solar panels for power generation. It is the deep space probes which require RTGs in order to generate power when the craft is far away from the sun.

 

[rootone]

Are both voyagers still phoning home ocassionally?
Last I heard they are still sending data, although at very low signal strength.

 

[SteamKing]

Are both voyagers still phoning home ocassionally?
Last I heard they are still sending data, although at very low signal strength.

The craft are capable of functioning until about 2020. After that time, due to reductions in available power, instruments will be shut down one by one, until the last instrument is deactivated about 2025.

https://en.wikipedia.org/wiki/Voyager_program

 

[RobertGC]

Thanks for responding. The nuclear fuel for nuclear thermal rockets is usually contained in rectangular canisters. I called them "fuel cells". I don't know what is the correct terminology.
They are different than just RTG's:

https://en.wikipedia.org/wiki/Nuclear_thermal_rocket

Bob Clark

 

[SteamKing]

Thanks for responding. The nuclear fuel for nuclear thermal rockets is usually contained in rectangular canisters. I called them "fuel cells". I don't know what is the correct terminology.
They are different than just RTG's:

https://en.wikipedia.org/wiki/Nuclear_thermal_rocket

Bob Clark

The critical distinction here is that all of these designs were paper studies or test prototypes. No actual spacecraft have been launched using this technology, and it is unlikely that any craft in the future will use it, because of radioactive contamination from the exhaust.

 

[russ_watters]

Thanks for responding. The nuclear fuel for nuclear thermal rockets is usually contained in rectangular canisters. I called them "fuel cells". I don't know what is the correct terminology.
They are different than just RTG's:

https://en.wikipedia.org/wiki/Nuclear_thermal_rocket

Bob Clark

Where did you get the idea that such a rocket would "only last for hours"? It looks to me like the limiting factor would be the reaction mass carried, not the nuclear fuel. So you could run them much longer if you wanted to (if it made sense to).

 

[RobertGC]

Here's one design that at full power only lasts a few hours:

Bimodal NTR.
Engine (Thrust Mode)
Thrust per engine 67,000 N
Total Thrust 200,000 N
T/Wengine 3.06
Exhaust Velocity 9,370 m/s
Specific Impulse 955 s
Propellant
Mass Flow 7.24 kg/s
Full Power
Engine Lifetime 4.5 hours
Reactor Power 335 MWthermal
http://www.projectrho.com/public_html/rocket/realdesigns.php#id--Bimodal_NTR

You can get it to last for months only by running it at much reduced power.

Bob Clark

 

[russ_watters]

Here's one design that at full power only lasts a few hours:

Bimodal NTR.
Engine (Thrust Mode)
Thrust per engine 67,000 N
Total Thrust 200,000 N
T/Wengine 3.06
Exhaust Velocity 9,370 m/s
Specific Impulse 955 s
Propellant
Mass Flow 7.24 kg/s
Full Power
Engine Lifetime 4.5 hours
Reactor Power 335 MWthermal
http://www.projectrho.com/public_html/rocket/realdesigns.php#id--Bimodal_NTR

You can get it to last for months only by running it at much reduced power.

Bob Clark

That link is really long and difficult to navigate, but I think you are misreading it. You need to find-out, specifically, what runs-out after 4.5 hours, because I'm pretty sure it is just the reaction mass (which they call "propellant"; liquid hydrogen), not the nuclear fuel.

 

[Astronuc]

Here's one design that at full power only lasts a few hours:

Bimodal NTR.
Engine (Thrust Mode)
Thrust per engine 67,000 N
Total Thrust 200,000 N
T/Wengine 3.06
Exhaust Velocity 9,370 m/s
Specific Impulse 955 s
Propellant
Mass Flow 7.24 kg/s
Full Power
Engine Lifetime 4.5 hours
Reactor Power 335 MWthermal
http://www.projectrho.com/public_html/rocket/realdesigns.php#id--Bimodal_NTR

You can get it to last for months only by running it at much reduced power.

Bob Clark

I don't think the core is limiting with respect to 4.5 hours. That may be sufficient for the mission. Then 4.5 hours can be a long time for very high temperatures.

For 4.5 hrs * 7.24 kg/s yields about 117.288 MT of propellant. That's quite a load - and that's just the propellant.

Usually direct thrust NTRs are fired for some period time, usually short compared to the mission duration, and the spacecraft coasts most of the way. One could coast into orbit at the destination, or if one travels faster, then one has to decelerate at the destination orbit, which costs more energy and propellant. The same then would occur on the return trip, if it is to be a round-trip.

The webpage has a comment "10 grams of Uranium-235 out of the 33,000 grams of ²³⁵U in each engine". Ten grams out of 33 kgU is not 0.03% FIMA which is not a lot of burnup, so the fission products would be insignificant. I'm not sure the source of the information, but perhaps it comes from the pdf in the link.

 

[Astronuc]

Also how can we improve the proportion of the nuclear fuel that actually gets consumed before we have to dispose of the fuel cell? I understand the fission reaction gets blocked by the fission products which inhibits the reaction from continuing. Couldn't we just have the reaction proceed in layers for a fuel from the outside in? When the outside fuel got consumed we would remove that and allow the inner layers to react. If we make the layers thin enough we should get little blockage from the fission-products.

Neutrons are highly penetrating, so pretty much all the fuel is exposed to neutrons and has fissioning going on. There is some shielding at the outer surface, for moderated systems where there is a moderator between fuel elements (rod or small spheres). The neutron flux falls off near the edges of the core, but we can use reflects to reduce the neutron leakage.

If we could get most of the fuel to react then we would need much less radioactive material to be sent to space so nuclear space propulsion would be much less controversial. Also if the ground based reactors could use most of the fuel, much less uranium would be needed to operate a reactor and you have lower operation costs and radioactive waste being produced.

In a critical system, we cannot get most of the fuel to react/fission, simply because there needs to be a critical mass after depletion of the excess fuel and accumulation of fission products. We could do coupled cores (gas-solid) or accelerator driven cores, but then one would have to address the problem of fission products, especially if gas-core expels fission products.

 

[RobertGC]

The problem of only a small proportion of the nuclear fuel actually being utilized is a common problem with nuclear reactors including ground based ones. Apparently the build up of fission products inhibits the fission process:

https://en.m.wikipedia.org/wiki/Burnup

But as Astronuc indicated it is particularly bad for space nuclear reactors. I'm trying to see if there are ways to improve that for the space reactors. And as indicated in the Wiki article it would be useful to improve it for the ground based reactors as well.

Bob Clark

 

[vemvare]

The "space reactor" you are referring to is a nuclear thermal rocket. Basically, a nuclear reaction heats ceramic fuel, which in turn heats hydrogen which is then expelled to create thrust. Considering design choices, two major differences between this and any other nuclear reactor is it is used to launch rockets and thereby it must have an extremely high power density (W/kg), also, it operates at temperatures much, much higher than even the Very High Temperature Reactor concept.

As it is supposed to be a part of a rocket, it uses a reaction mass which is also what keeps the rocket from overheating. This is hydrogen, and it runs out within a few minutes of operation. Since the reactor must be designed to handle the two extreme parameters mentioned before, and runs out of fuel anyway, the ability for the system to operate continuously for a long time isn't important.

In a nuclear reactor designed to be used in space as opposed to a planetary surface the nuclear reactions would behave exactly the same as they would on Earth, but if it was supposed to run for a longer period of time the only possible "coolant" would be radiative cooling. This would make a higher output (and input) temperature highly beneficial, but the rest of the very difficult discussion is where the limits of materials technology takes this. There would be no need however for the reactor-system to have the insane power density (hundreds of kW/kg) or temperatures of a nuclear thermal rocket.

 

[QuantumPion]

The problem of only a small proportion of the nuclear fuel actually being utilized is a common problem with nuclear reactors including ground based ones. Apparently the build up of fission products inhibits the fission process:

https://en.m.wikipedia.org/wiki/Burnup

But as Astronuc indicated it is particularly bad for space nuclear reactors. I'm trying to see if there are ways to improve that for the space reactors. And as indicated in the Wiki article it would be useful to improve it for the ground based reactors as well.

Bob Clark

You misinterpreted Astronuc's post to the exact opposite conclusion. Burnup is completely irrelevant for a NTR. The rocket is only limited by the amount of hydrogen propellent you can carry, the reactor itself could run for months or even years. I'm sure it is more limited by the materials the reactor is made out of wearing out than running out of reactivity in the fuel.

The main problem preventing NTR's from practical use is the (public perception of the) hazard of a launch failure and spreading radioactive contamination. Which wouldn't be that bad since the reactor would not be activated until it's in orbit, so the worse case is fresh UO2 spread out over a wide area. But try explaining that to greenpeace...

 

[RobertGC]

I think Vemvare explained the situation correctly. The Nuclear Thermal Rockets (NTR's) need much higher temperatures than ground based reactors, as much as three times higher. That causes increased destruction of the various components of the fuel canisters.
Or said another way the ground based reactors operate at lower power than they are really capable thereby extending the fuel canister lifetimes. This is the same way that the some space reactors, such as the bimodal, will extend their lifetimes for low power, i.e., power generation, non-propulsive, long life operation.
If a way could be found to have the space reactors operate at lower temperatures while at the same time deliver the reaction mass, usually hydrogen, at high speed, then we would have long operation times for the propulsion system while still at high Isp.

Bob Clark

 

[Astronuc]

The accumulation of fission products is inherent in the use of nuclear fuel. Every fission event produces two atoms, or infrequently, three atoms (ternary fission). About 30% of fission products are Xe, Kr isotopes, and another percentage is comprised of volatiles like Cs, Te, I, Br, . . . Accumulation of fission products causes the fuel material to swell, and the fission products will compete with the fuel for neutrons.

Radiation effects to solid materials are also inherent in nuclear reactor system, and these are also design considerations. At high temperatures, radiation damage to crystal structures are annealed, and fuel creep (both thermal and irradiation-induced) must be considered with respect to structural/geometric stability. Also, NTR cores have to be designed with the consideration of highly turbulent propellant flow - so fatigue and erosion are concerns.

Rocket motors necessarily operate at high temperature as much as possible, but this means swelling and creep of the fuel components. The fuel for nuclear thermal rockets is most likely carbides of U, Pu, or mixed carbides like (U,Zr)C. On the other hand, for power generation, a nuclear thermal core would be operated at lower power density/temperature depending on the design of the thermodynamic cycle (e.g., Brayton cycle).

LWR fuel uses UO₂, and such fuel is operated with strict limits on fuel temperature, cladding strain (< 1% total in the event of a transient) and rod internal pressure. The fuel designer has to design fuel subject to those limits.

If a way could be found to have the space reactors operate at lower temperatures while at the same time deliver the reaction mass, usually hydrogen, at high speed, then we would have long operation times for the propulsion system while still at high Isp.

NTR motors are relatively low I_sp compared to nuclear electric systems. Usually, thrust decreases with I_sp (due to low mass flow rate) and power demands increase.

 

[russ_watters]

If a way could be found to have the space reactors operate at lower temperatures while at the same time deliver the reaction mass, usually hydrogen, at high speed, then we would have long operation times for the propulsion system while still at high Isp.

Use the reactor to generate electricity for ion propulsion. You'll get much higher impulse out of the reaction mass.

 

[snorkack]

The accumulation of fission products is inherent in the use of nuclear fuel. Every fission event produces two atoms, or infrequently, three atoms (ternary fission). About 30% of fission products are Xe, Kr isotopes, and another percentage is comprised of volatiles like Cs, Te, I, Br, . . . Accumulation of fission products causes the fuel material to swell, and the fission products will compete with the fuel for neutrons.

It is inherent in the use of solid nuclear fuel, where fission products get stuck.
A reactor based on fluid fuel might have online chemical separation of wastes from fuel, allowing high burnups.
However, no working fluid fuel reaction seems to have operated so far. And even if one were developed, it might be less suitable for space use if it turns out to face inconvenient space or weight constraints, rely on gravity or something else such.

 

[Astronuc]

It is inherent in the use of solid nuclear fuel, where fission products get stuck.
A reactor based on fluid fuel might have online chemical separation of wastes from fuel, allowing high burnups.
However, no working fluid fuel reaction seems to have operated so far. And even if one were developed, it might be less suitable for space use if it turns out to face inconvenient space or weight constraints, rely on gravity or something else such.

Then there would be the matter of transferring heat (thermal energy) from the liquid fuel system to the hydrogen propellant. A heat exchanger could be problematic, and much less effective than transferring heat from solid fuel to liquid/gaseous propellant. Adding a chemical separation to the system would add another layer of complexity, as well as mass not related to propulsion.

 

[mheslep]

However, no working fluid fuel reaction seems to have operated so far.

4 yrs of operation
https://en.wikipedia.org/wiki/Molten-Salt_Reactor_Experiment

 

[vemvare]

I think we're talking in different directions here; a nuclear thermal rocket (NTR) is not the same thing as a nuclear-electric propulsion (NE) system. Considering the dangers involved in using an NTR to ferry something to orbit, adding the fact that acceleration is much less important than specific impulse for orbit-to-orbit transfers and that NE whether it is powering ion engines, vasimr or nanofet would have a superior specific impulse, the NTR concept to me makes no sense.

 

[mheslep]

NTR concept to me makes no sense.

Yes, nor does NE make any sense for lift to orbit.