Nuclear Thermal Rocket questions (riding around space in a uranium powered steamboat)

[PTNLemay]

Quick Explanation:

From my limited understanding of NTRs, the usual limitation placed on how efficient it can be is how hot you can make the engine without it melting. Hotter the engine, the faster the exhaust, the more thrust per unit of fuel. Some people have suggested using liquid-fuelled cores (or even the gaseous cores) but those have the frustrating tendency of leaving a long-lived radioactive exhaust trail behind your ship.

Thought Experiment:

So I'm thinking... what if we have an imaginary substance, something that can withstand the insane temperatures and pressures of liquid uranium (say a material with a melting point of 5,000 degrees). We pour the uranium into these tubes and seal them, it can then fission away and stay contained, while emitting electromagnetic radiation (maybe as high as gamma) out to heat the propellent. Ideally I'd want it to be hot enough that it could use water as a propellent and still achieve a good exhaust velocity. That way we wouldn't need to resort to awkward cryogenic propellent like liquid hydrogen. We'd be able to "refuel" easily.

My Question:

What I really want to know is if we did do this, how dangerous would the uranium deposit be? I'm not too worried about the gamma rays, I mean the crew will be shielded from cosmic rays anyway. But would increasing it's temperature that high make it more at risk of hitting super-criticality, or less? I'm all for imagining a super-material that can withstand crazy temperatures, but even I acknowledge you can't ignore a chain reaction if it results in a nuclear explosion.

I looked around these articles but couldn't find an answer. Anyone here able to help me out?

http://en.wikipedia.org/wiki/Critical_mass

http://en.wikipedia.org/wiki/Criticality_accident

Edited September 13, 2013 by PTNLemay
typo

[Kryten]

The idea you're proposing has already been thought of, the closed-cycle gas-core NTR (aka 'Nuclear lightbulb'). Usually the material proposed is quartz. And no, the temperature of the reactor doesn't effect reaction rate (and so stuff like chance of prompt criticality) unless you're talking millions rather than thousands of degrees.

[SargeRho]

How are NTRs sci-fi theory?

There is at least one NERVA ROVER engine stuffed away in a hangar somewhere.

They explosively tested the engine reactors. Apparently they were sturdy enough to not spray radiation all over the place.

[lajoswinkler]

What's this "uranium deposit" you're talking about?

You are aware that metallic uranium is never used in reactors?

[Kryten]

You are aware that metallic uranium is never used in reactors?

It's the usual choice for gas-core, which is what's he's talking about here.

[lajoswinkler]

Isn't that only theoretical, and even then, hexafluoride is used more often?

The notion of gaseous metal undergoing fission kind of freaks me out.

[Mr Shifty]

And no, the temperature of the reactor doesn't effect reaction rate (and so stuff like chance of prompt criticality) unless you're talking millions rather than thousands of degrees.

I don't know anything about gas core reactors, but in the PLWRs I am familiar with, the core temperature affects reactivity because it changes the temperature (and thus density) of the moderator. Doesn't the gas core nuclear rocket use H2 as a propellant and moderator? Wouldn't the density of the gaseous H2 be highly dependent on the temperature in the core?

[Kryten]

The kind of core he's talking about can't use fuel as moderator, as it's not in direct contact with the core. In all honestly I've no idea how closed-cycle cores are supposed to be moderated. EDIT: Looking at a few concepts, none of them make any mention of moderation. They all also seem to use very high-grade fuel, so I think they're just intended to be 'fast' reactors.

Edited September 13, 2013 by Kryten

[Bunsen]

I think Shifty is right about the hydrogen serving as the moderator, at least in part. The reaction would be influenced by neutrons that leave the core and get reflected back, and hydrogen flow would have a large influence on their spectrum. Since thermal, or at least partially thermalized, neutrons are disproportionately effective compared to the intra-core fast neutrons, I expect that much of the reactivity control could be accomplished with external reflectors and controlled hydrogen flow.

[TheSaint]

I don't know anything about gas core reactors, but in the PLWRs I am familiar with, the core temperature affects reactivity because it changes the temperature (and thus density) of the moderator. Doesn't the gas core nuclear rocket use H2 as a propellant and moderator? Wouldn't the density of the gaseous H2 be highly dependent on the temperature in the core?

I think the idea is that you design the core so that the reaction mass provides some moderation, but not enough to sustain criticality, and then you have some type of high-temperature neutron reflector/moderator (like a beryllium-laced ceramic or something similar) that can be moved to cover all, some, or none of the core. When you need to change the reactivity in the core, such as when you are shutting down or changing your thrust levels, you simply move the neutron reflector accordingly. When you shut down you keep a bleed of hydrogen around the core to remove decay heat.

Edited September 13, 2013 by TheSaint

[Mr Shifty]

The kind of core he's talking about can't use fuel as moderator, as it's not in direct contact with the core. In all honestly I've no idea how closed-cycle cores are supposed to be moderated. EDIT: Looking at a few concepts, none of them make any mention of moderation. They all also seem to use very high-grade fuel, so I think they're just intended to be 'fast' reactors.

Yeah, makes sense. I'm curious how the reaction is contained, i.e. what keeps the engine from becoming a bomb. The concept papers I've read seem ludicrous: a vortex of neon gas to keep the uranium plasma from touching anything. The fused silica glass has to be kept at a particular temperature or it becomes opaque, absorbs too much radiation, and melts. It's hard to imagine how this could be made remotely safe enough to operate.

[Mr Shifty]

I think the idea is that you design the core so that the reaction mass provides some moderation, but not enough to sustain criticality, and then you have some type of high-temperature neutron reflector/moderator (like a beryllium-laced ceramic or something similar) that can be moved to cover all, some, or none of the core. When you need to change the reactivity in the core, such as when you are shutting down or changing your thrust levels, you simply move the neutron reflector accordingly. When you shut down you keep a bleed of hydrogen around the core to remove decay heat.

OK. That makes sense. Thanks for the explanation.

[PTNLemay]

Usually the material proposed is quartz.

Quartz is alright... but it's brittle, if you make the entire core out of a quartz shell one good bump risks cracking it. And it's melting point is actually still quite low. ~1600 C according to wikipedia. It's good if we want a scenario that's strictly realistic, but I want to push a bit into imaginary here. Get an NTR that really has a killer exhaust velocity (and an equally robust thrust). In real life we can get great thrust or great exhaust velocity, but rarely both. The only times it might work is with excessive use of nuclear power with things like orion boom-boom rockets, and like I said earlier, I dislike the radioactive waste that would produce. So I want a technology that will have all the might of nuclear, but contain all of the waste.

How are NTRs sci-fi theory?

There is at least one NERVA ROVER engine stuffed away in a hangar somewhere.

They explosively tested the engine reactors. Apparently they were sturdy enough to not spray radiation all over the place.

As I said in the first post, the idea here is to suppose a material that would be able to handle ~5000 degrees without weakening. NTRs exist of course, but they can't be heated much more than conventional nuclear reactors can (~500 degrees I think). This means that the propellent being shot out of a moden NTR (if we were still building them using modern materials) has a fraction of the true potential energy it could have imparted onto it by the nuclear fuel.

What's this "uranium deposit" you're talking about?

You are aware that metallic uranium is never used in reactors?

I imagine that in order to get up to 5000 degrees we would need a very high concentration. I figure metallic uranium would fit the bill. But even if it doesn't... I still refer to Uranium as the fuel because that's where the energy primarily comes from.

In all honestly I've no idea how closed-cycle cores are supposed to be moderated.

Right, my question is what would happen if a nuclear thermal rocket was allowed to heat up to insane temperatures (by being unmoderated I guess) but be incapable of melting. Would it eventually explode?

Edited September 13, 2013 by PTNLemay

[drakesdoom]

I have never understood the focus on saftey. We are orbiting the galaxy at 220km/s along with an open chain fusion reaction with just distance and a very small amount of air as shielding.

Honestly after that what is dangerous?

[PTNLemay]

Oh and to clarify, the propellent is going to be water. I know it's exhaust velocity won't be as good as H2, but it will store more easily. The idea is to increase the exhaust velocity up to satisfactory speeds by making the engine run hotter and dump more energy into the remass.

@ Drake

I don't mind the radiation itself, like I said earlier there's already plenty of cosmic radiation out there. What worries me is radioactive waste that gets dumped when you use an open gas-core design. Those are more like... toxic by-products of the nuclear radiation. If we imagine humanity using these vehicles en-mass, over a long period of time the Earth would get filled with a kind of radioactive "smog". Which is good if you want to imagine a gritty post-apocalyptic future Earth, but that's not what I'm envisioning here.

Also the open design is wasteful. I imagine you'd have to refuel the fissible material almost as often as you have to refuel on water. With a closed design I imagine the fuel could endure more similarly to how it does in modern nuclear vessels (with lifespans measured in decades).

[Mr Shifty]

Right, my question is what would happen if a nuclear thermal rocket was allowed to heat up to insane temperatures (by being unmoderated I guess) but be incapable of melting. Would it eventually explode?

Moderators make the reaction more effective by slowing down the fission-produced neutrons so that they stay in the core. The discussion above has been speculation that these reactors would not be able to achieve criticality (criticality is good -- means the reaction is exactly self-sustaining) using only the fast neutrons produced by fission. You'd make the reactor critical by lowering a reflector or introducing a moderating gas into the chamber surrounding the core which would reflect enough neutrons back into the core to sustain the reaction. To shut it down, you remove the reflector. NB: The melting point of uranium is about 1100 degrees centigrade. See the discussion on this page for how the thermodynamics were supposed to work out:

http://up-ship.com/blog/?p=6694

[Mr Shifty]

I have never understood the focus on saftey. We are orbiting the galaxy at 220km/s along with an open chain fusion reaction with just distance and a very small amount of air as shielding.

Honestly after that what is dangerous?

The problem is that you spend $200 million on just launch costs, disregarding the cost of developing the payload and all the millions of dollars of ancillary costs associated with the mission. A shame if you get your nuclear rocket into orbit, turn it on, and the whole thing melts into a glowing ball of liquid radioactive metal orbiting the Earth... until it re-enters the atmosphere. Not sure if it would pose a significant health hazard at that point (my intuition leans toward no), but it would certainly be a major political boondoggle and a potential giant step backwards for space exploration.

[TheSaint]

Quartz is alright... but it's brittle, if you make the entire core out of a quartz shell one good bump risks cracking it. And it's melting point is actually still quite low. ~1600 C according to wikipedia. It's good if we want a scenario that's strictly realistic, but I want to push a bit into imaginary here. Get an NTR that really has a killer exhaust velocity (and an equally robust thrust). In real life we can get great thrust or great exhaust velocity, but rarely both. The only times it might work is with excessive use of nuclear power with things like orion boom-boom rockets, and like I said earlier, I dislike the radioactive waste that would produce. So I want a technology that will have all the might of nuclear, but contain all of the waste.

As I said in the first post, the idea here is to suppose a material that would be able to handle ~5000 degrees without weakening. NTRs exist of course, but they can't be heated much more than conventional nuclear reactors can (~500 degrees I think). This means that the propellent being shot out of a moden NTR (if we were still building them using modern materials) has a fraction of the true potential energy it could have imparted onto it by the nuclear fuel.

I imagine that in order to get up to 5000 degrees we would need a very high concentration. I figure metallic uranium would fit the bill. But even if it doesn't... I still refer to Uranium as the fuel because that's where the energy primarily comes from.

Right, my question is what would happen if a nuclear thermal rocket was allowed to heat up to insane temperatures (by being unmoderated I guess) but be incapable of melting. Would it eventually explode?

Most designs actively cool the quartz light bulb with the hydrogen reaction mass, so it can take a lot higher temperature than the actual melting point of the quartz. The number I see most commonly quoted is about 25,000C.

The temperature of the fuel doesn't have a huge effect on the amount of energy it produces. Remember, this is a nuclear reaction, not a chemical reaction. If anything, as the fuel heats up it will expand, which will reduce the density of the fuel and reduce the rate of fission.

And, for the record, the NERVA rockets back in the 60s operated around 1,700C.

[PTNLemay]

@ Saint

But isn't there a direct proportion to the energy being transfered to the propellant? Hotter core, more energy, faster expanding remass.

Another way of looking at what I'm trying to imagine here is an NTR that uses water as a propellent, but that's energetic enough that the exhaust velocity would match (or exceed) liquid hydrogen. And for that I'm willing to break a bit of realism rules by using super materials. I like H2 and all that, but I find it annoying that it has to be cryogenically stored, and even when it is, it escapes over time. That makes it very difficult to use in an in-situ rerource utilization situation. With water you can just find some ice, melt it, and filter the resulting water.

@ Shifty

Thanks for the link. It does appear to be a bit... beyond my knowledge, to be honest (especially those diagrams). Also it runs into the usual limitations of realistic materials, namely that everything we use has trouble holding it's shape at these temperatures and wants to melt like butter sitting under a blow-torch. I don't want to ignore real science, but I don't want to be anchored by it either.

Edited September 13, 2013 by PTNLemay

[TheSaint]

@ Saint

But isn't there a direct proportion to the energy being transfered to the propellant? Hotter core, more energy, faster expanding remass.

Another way of looking at what I'm trying to imagine here is an NTR that uses water as a propellent, but that's energetic enough that the exhaust velocity would match (or exceed) liquid hydrogen. And for that I'm willing to break a bit of realism rules by using super materials. I like H2 and all that, but I find it annoying that it has to be cryogenically stored, and even when it is, it escapes over time. That makes it very difficult to use in an in-situ rerource utilization situation. With water you can just find some ice, melt it, and filter the resulting water.

@ Shifty

Thanks for the link. It does appear to be a bit... beyond my knowledge, to be honest (especially those diagrams). Also it runs into the usual limitations of realistic materials, namely that everything we use has trouble holding it's shape at these temperatures and wants to melt like butter sitting under a blow-torch. I don't want to ignore real science, but I don't want to be anchored by it either.

You almost certainly can design a NTR to use water as a fuel. Just realize that water will have different neutron moderation characteristics than liquid hydrogen, so the propellants probably won't be interchangeable. And it will be slightly less efficient than a liquid hydrogen NTR operating at the same temperature.

If you're going to wander off the reality reservation you can declare that your light bulb can withstand any temperature you desire. Bi-phase carbide, adamantium, scrith, unobtainium, call it whatever you want. Just realize that the primary heat transfer from a gas core is through the radiation of hard ultraviolet light, so whatever material you declare it to be made out of will have to be transparent in that wavelength.

[Temstar]

Water is tricky to use, at those temperatures water molecule will be cracked apart into oxygen and hydrogen atom by the high temperate. This is good for Isp because the resulting gas is greatly expanded in volume but it's very hard on the engine because super high temperature oxygen ion is going to, well, oxidise anything it comes in contact with, it be the next hydrogen it runs into or the metal walls of your engine.

[TheSaint]

Water is tricky to use, at those temperatures water molecule will be cracked apart into oxygen and hydrogen atom by the high temperate. This is good for Isp because the resulting gas is greatly expanded in volume but it's very hard on the engine because super high temperature oxygen ion is going to, well, oxidise anything it comes in contact with, it be the next hydrogen it runs into or the metal walls of your engine.

Not terribly hard to deal with. Just ensure that the surfaces in your combustion chamber are made of (or clad with) highly corrosion resistant materials. Monel, Hastelloy, Inconel, all sorts of solutions there.

[PTNLemay]

In an ionized gas, the electrons are basically being stripped, right? So in a really really high temperature environment the oxygen would be losing a lot of it's valence electrons. The excess electrons in the metal & alloys of the engine-casing would fill those electron gaps, and through that we'd get oxidization (this is me thinking out loud, please correct me if/when I make a mistake in my reasoning).

Isn't there a way to apply an electric current into the metal casing to help provide additional electrons and inhibit oxidization? I remember hearing that there are already such devices used on boats and trucks and stuff like that. If we assume sufficiently advanced technology, maybe humanity will have invented really fancy ones that they put on their rocketships.

[DaveofDefeat]

You can always coat your engine in something other than metal.

[lajoswinkler]

Extremely hot water will attack most metals even if it molecules are still intact, so superhot water that dissociates into atoms will eat even more stuff.