Nuclear salt-water rocket propulsion

[lajoswinkler]

The radioactivity of the exhaust isn't an issue for two reasons.

1) The exhaust is moving so fast that it won't stick around long enough to cause any issues.

2) The exhaust is moving so fast that most of the nuclei in it will have more kinetic energy than your average alpha particle. Anything that gets in the way of the exhaust will be utterly obliterated.

1) The nozzle is hot, permitting fission products infusion. The system would become insanely contaminated with extremely radioactive elements in a short amount of time.

2) I'm not so sure about the kinetic energy, but even if it was true, it would not be that dreadful. It's somewhat worse than standing in front of an ion engine in vacuum. if we're talking about the instant damage and ignore the fact you'd get huge doses of radiation that would kill you in a few days.

In addition to all of the other downsides listed above, also consider this: One micrometeoroid strike on your fuel tank, one fuel leak, one throttle failure, or one of countless other possible mechanical failures could release enough fuel to start a chain reaction, which would release more fuel, which would make you bunkmates with an uncontrolled nuclear explosion. I wouldn't ride in the damn thing.

No. Just no.

Compromising the structural integrity of the container that holds the reaction mass (which acts as a coolant, too... at least partially) would result in a meltdown, unless other safety systems would turn on, something I don't see a reason against. Meltdowns happen because of the fission products that decay and produce heat.

You'd end up with a glowing lava blob where your ship's propulsion system was, and it would be spectacular. But no nuclear explosions.

Unfortunately, no. Nuclear weapons are frighteningly easy to design. Remember, the Manhattan Project didn't have access to computers, they designed all of their weapons (including the very complex implosion weapons) with pencil, paper, and slide rule. This is why non-proliferation efforts stopped focusing on design knowledge and are instead now focused on limiting access to nuclear material. And even that is failing miserably. We need to stop trying to prevent folks from getting nuclear weapons and instead try to figure out how we'll keep the world safe when everyone has nuclear weapons.

Frighteningly easy to make? Manhattan project had an army of people to do the calculations. People with slide rules and computers (females doing calculations; the original meaning of the term). Even with all that, there were occasions of bombs fizzling instead of detonating.

It's frighteningly tough to make a nuclear bomb from the scratch. This has already been discussed elsewhere on KSP forums.

And regarding the safety, it's the mutually asserted destruction that's keeping everyone from using them.

[DerekL1963]

Frighteningly easy to make? Manhattan project had an army of people to do the calculations. People with slide rules and computers (females doing calculations; the original meaning of the term). Even with all that, there were occasions of bombs fizzling instead of detonating.

And today, you can look up all the stuff they had to figure out from scratch on the internet. The problems they worried about that could cause fizzles are well covered in the open literature. Consider what NASA had to go through to make that first suborbital flight... and then look at Scaled Composites and Copenhagen Suborbitals. Just because it took armies of workers and the absolute brightest of techs and scientists fifty years does not mean the same holds true today.

It's frighteningly tough to make a nuclear bomb from the scratch. This has already been discussed elsewhere on KSP forums.

Not by anyone with any knowledge of the issue apparently. It's frightfully easy, much easier than you seem to think.

As TheSaint said, there's a reason why non proliferation efforts have shifted to controlling access to the materiel.

[-Velocity-]

Frighteningly easy to make? Manhattan project had an army of people to do the calculations. People with slide rules and computers (females doing calculations; the original meaning of the term). Even with all that, there were occasions of bombs fizzling instead of detonating.

It's frighteningly tough to make a nuclear bomb from the scratch. This has already been discussed elsewhere on KSP forums.

And regarding the safety, it's the mutually asserted destruction that's keeping everyone from using them.

An "army of workers" was used to create the industry to produce nuclear material. All you need for the most basic nuclear bomb, a gun-type using U-235, is a ring of U-235, and a spike of U-235 over which that ring can fit. Oh, and you need to know how much a critical mass is, but that's public knowledge.

Just shoot your ring of U-235 down a barrel with high explosives, and on impact have it fit around the spike, and now you have a supercritical mass, fully assembled before it can fizzle itself apart. Next comes a BIG boom.

In fact, part of what this "army of workers" did was work on an alternate nuclear weapons design, the implosion type, which uses explosive lenses to compress a subcritical sphere of plutonium down to a size small enough to go supercritical. This was necessary for plutonium because plutonium-239 is contaminated with shorter half-life isotopes, the neutron emissions from which will invariably cause a gun-type nuclear weapon to fizzle.

The gun-type nuclear weapon design was considered so basic and foolproof a nuclear version of it was never even tested before it was dropped on Hiroshima. However, the scientists had to test the implosion type first before they used it, just to make sure (the Trinity test), and then the other implosion weapon that the US had was dropped on Nagasaki.

Tell me, if nuclear weapons are so hard to make work, HOW COMES THE DESIGN OF THE FIRST NUCLEAR BOMB EVER USED IN WAR WAS NEVER EVEN TESTED BEFORE IT WAS DROPPED?

Think on that a minute.

Edited September 12, 2013 by |Velocity|

[TheSaint]

No. Just no.

Compromising the structural integrity of the container that holds the reaction mass (which acts as a coolant, too... at least partially) would result in a meltdown, unless other safety systems would turn on, something I don't see a reason against. Meltdowns happen because of the fission products that decay and produce heat.

You'd end up with a glowing lava blob where your ship's propulsion system was, and it would be spectacular. But no nuclear explosions.

The whole point of the NSWR is that when the fuel/reaction mass leaves the safe configuration of the storage tanks and enters the combustion chamber it undergoes nuclear fission. If it leaves the safe configuration of the storage tanks and enters some other space through a leak then it still has the potential to undergo nuclear fission. And even a small amount of nuclear fission near the storage tanks or other fuel handling equipment would have the potential to bombard the fuel in the storage with a large number of neutrons, not to mention further damage to the safe configuration of the storage tanks themselves, leading to a criticality accident. Otherwise known as a BOOM (look up SL-1 for a story about what happens during a criticality accident). Sorry, still wouldn't ride in the damn thing.

[magnemoe]

An "army of workers" was used to create the industry to produce nuclear material. All you need for the most basic nuclear bomb, a gun-type using U-235, is a ring of U-235, and a spike of U-235 over which that ring can fit. Oh, and you need to know how much a critical mass is, but that's public knowledge.

Just shoot your ring of U-235 down a barrel with high explosives, and on impact have it fit around the spike, and now you have a supercritical mass, fully assembled before it can fizzle itself apart. Next comes a BIG boom.

In fact, part of what this "army of workers" did was work on an alternate nuclear weapons design, the implosion type, which uses explosive lenses to compress a subcritical sphere of plutonium down to a size small enough to go supercritical. This was necessary for plutonium because plutonium-239 is contaminated with shorter half-life isotopes, the neutron emissions from which will invariably cause a gun-type nuclear weapon to fizzle.

The gun-type nuclear weapon design was considered so basic and foolproof a nuclear version of it was never even tested before it was dropped on Hiroshima. However, the scientists had to test the implosion type first before they used it, just to make sure (the Trinity test), and then the other implosion weapon that the US had was dropped on Nagasaki.

Tell me, if nuclear weapons are so hard to make work, HOW COMES THE DESIGN OF THE FIRST NUCLEAR BOMB EVER USED IN WAR WAS NEVER EVEN TESTED BEFORE IT WAS DROPPED?

Think on that a minute.

Its hard to make weapon grade uranium. As you say you needed major infrastructure to make it, plutonium is made in reactors so its easier.

However an U-235 bomb is simple, you need an neutron source too but that is not hard.

But the main reason why the U-235 bomb was not tested was not at it was an so simple design but that they did not have U-235 to perform an test and the bomb would probably work.

In short an plutonium bomb use cheaper materials but is harder to make, the plutonium bomb is lighter but more bulky so the nuclear artillery grenades was uranium bombs.

The weight might be one reason why we don't use uranium bombs anymore, another reason is that I think its simpler to expand it into an hydrogen bomb.

Finally the uranium bomb has one serious drawback, if the plane carrying it crashes the bomb might go off. An plutonium bomb require all the triggers around the bomb to go off at once and this will not happen in an accident.

[lajoswinkler]

An "army of workers" was used to create the industry to produce nuclear material. All you need for the most basic nuclear bomb, a gun-type using U-235, is a ring of U-235, and a spike of U-235 over which that ring can fit. Oh, and you need to know how much a critical mass is, but that's public knowledge.

Just shoot your ring of U-235 down a barrel with high explosives, and on impact have it fit around the spike, and now you have a supercritical mass, fully assembled before it can fizzle itself apart. Next comes a BIG boom.

In fact, part of what this "army of workers" did was work on an alternate nuclear weapons design, the implosion type, which uses explosive lenses to compress a subcritical sphere of plutonium down to a size small enough to go supercritical. This was necessary for plutonium because plutonium-239 is contaminated with shorter half-life isotopes, the neutron emissions from which will invariably cause a gun-type nuclear weapon to fizzle.

The gun-type nuclear weapon design was considered so basic and foolproof a nuclear version of it was never even tested before it was dropped on Hiroshima. However, the scientists had to test the implosion type first before they used it, just to make sure (the Trinity test), and then the other implosion weapon that the US had was dropped on Nagasaki.

Tell me, if nuclear weapons are so hard to make work, HOW COMES THE DESIGN OF THE FIRST NUCLEAR BOMB EVER USED IN WAR WAS NEVER EVEN TESTED BEFORE IT WAS DROPPED?

Think on that a minute.

You're wrong about the army of workers. There was a large number of people doing the calculations for the Trinity, as well as lots of people working on refining the material. The famous women turning dials of the centrifuges, not knowing they're participating in the making of a horrible weapon are on this photo.

But the calculations for the fission took the most human resources. Hordes of people with slide rules.

Are you aware of the haste of the whole project? It was literally a speedy work to make the bomb before the Nazis make one, and there was not enough material to make tests. The refinement techniques were primitive, so the highly enriched uranium-235 was extremely precious. That's why they did so much calculations.

You can't just shove to subcritical pieces of uranium together and expect a nuclear detonation. That's not how it works. It will fizzle.

In later years, with the onset of the Cold war, refinement techniques became a lot better, so more material could be made, and so the era of obsessive testing began.

The whole point of the NSWR is that when the fuel/reaction mass leaves the safe configuration of the storage tanks and enters the combustion chamber it undergoes nuclear fission. If it leaves the safe configuration of the storage tanks and enters some other space through a leak then it still has the potential to undergo nuclear fission. And even a small amount of nuclear fission near the storage tanks or other fuel handling equipment would have the potential to bombard the fuel in the storage with a large number of neutrons, not to mention further damage to the safe configuration of the storage tanks themselves, leading to a criticality accident. Otherwise known as a BOOM (look up SL-1 for a story about what happens during a criticality accident). Sorry, still wouldn't ride in the damn thing.

It's still not a nuclear explosion. There is a strict, fixed definition of what that is.

I wouldn't be happy to ride it, either.

[Temstar]

Its hard to make weapon grade uranium. As you say you needed major infrastructure to make it, plutonium is made in reactors so its easier.

That's the general idea, but no it's like this:

For an uranium bomb the isotope you need is U-235. Natural uranium is 0.7% U-235 with most of the rest U-238. To make a bomb you need weapons grade uranium which means around 90% concentrated U-235. To separate U-235 from U-238 is extremely difficult because they are the same element and so have the same chemical properties. You cannot use standard chemistry methods to separate isotopes, hence the use of a huge numbers of centrifuges to spin the two isotopes apart.

For plutonium bomb you need Pu-239. When you breed plutonium in a reactor you create both Pu-239 and Pu-240 together. You cannot separate Pu-239 from Pu-240 with any method!, at least not with any practical method. However you can control the Pu-239 to Pu-240 ratio by controlling how long your fuel remains in the reactor. Once you're done the breeding you remove the nuclear fuel, use normal chemical methods to get rid of any other elements and the Pu that you get is what you have to work with. The higher the Pu-240 the more likely your bomb will fizzle instead of go boom.

However an U-235 bomb is simple, you need an neutron source too but that is not hard.

You don't need neutron source for either type of bombs, the fuel itself is radioactive so is always decaying slowing and producing neutron. That's your built in neutron source.

But the main reason why the U-235 bomb was not tested was not at it was an so simple design but that they did not have U-235 to perform an test and the bomb would probably work.

The U-235 bomb (Little Boy) wasn't tested because it contained the entire world's supply of U-235 at time time in its core! There was literally no more U-235 around for the time being anywhere in the world to build another test bomb. However since that U-235 is enriched to weapons grade the scientist know that it won't fizzle. The U-235 is pure enough that you simple shoot a smaller piece of U-235 into a bigger piece with a hole in the middle with what basically is a small cannon inside the bomb and it will go off nuclear. With the plutonium bomb, because the reactor breed plutonium is contaminated with Pu-240 if you use the same simple design the plutonium will blow itself apart in a fizzle as soon as the little piece get near the big piece without there being time for the pieces complete the assembly. Initially the plan to deal with this is to use a longer cannon to shoot the plutonium piece so it flies into the target faster (the "Tall Man" design). But even this method wouldn't be able to complete assemble of the core fast enough to prevent a fizzle.

So the final plan is what's called the "Fat Man" design. Basically you have a ball of plutonium with a hollow centre. You then completely surround this ball with a shell of high explosive. When you want the bomb to go off you set off all the high explosives at the same time and the shockwave compresses the plutonium ball so that the hollow centre collapses and the plutonium is squeezed into a much smaller solid sphere. This sphere will then be above critical and go off. This method of assembling a core is much more efficient (and safer) than the gun method of Little Boy since a sphere is the ideal shape (lowest surface to volume ratio for all shapes, so it keeps the neutrons inside for longer) and by squeezing it from all side you get a much much higher density of the core material. This method is so good that nowadays all bomb use this design, even if for some reason its fuel is U-235.

In short an plutonium bomb use cheaper materials but is harder to make, the plutonium bomb is lighter but more bulky so the nuclear artillery grenades was uranium bombs.

The weight might be one reason why we don't use uranium bombs anymore, another reason is that I think its simpler to expand it into an hydrogen bomb.

Both types of bombs weigh the same. Plutonium bombs require a lot better knowledge of nuclear physics and explosive lenses control in order to work properly. But once you do know how to make plutonium bomb the actual plutonium is easier to breed than it is to enrich natural uranium. So that's why all the big nuclear weapon states use plutonium bombs. On the other hand for countries like Iran building their first bombs they don't have enough hands on knowledge to be able to build a plutonium bomb with Fat Man design, so they go for Little Boy style uranium bombs because it's a safer bet, even if producing weapons grade uranium is way more involved.

Finally the uranium bomb has one serious drawback, if the plane carrying it crashes the bomb might go off. An plutonium bomb require all the triggers around the bomb to go off at once and this will not happen in an accident.

That's only for the actual Little Boy itself, later uranium bomb also use the Fat Man design so they don't have this problem. It's a problem with the particular design rather than the fission fuel itself.

Edited September 12, 2013 by Temstar

[X10Z]

Not really an explosion per se, but a continuous release of energy. Much more dangerous is the possibility of meteorite impacts destroying your tanks and the uranium gaining critical mass in places you don't want it to.

[X10Z]

Besides, wouldn't any rocket using it going to use rcs and conventional rockets for rendezvous and docking?

[magnemoe]

Not really an explosion per se, but a continuous release of energy. Much more dangerous is the possibility of meteorite impacts destroying your tanks and the uranium gaining critical mass in places you don't want it to.

Its an very energetic reaction who is itself is pretty dangerous.

It would also require an pretty uniform flow of plutonium salt into the reaction chamber.

You would not need an meteorite impact for things to go wrong. any leak would be bad, you would also get structural problems because of shaking if your reaction rate was uneven.

The design make an orion engine look safe and nice.

[SciMan]

To put the amount of raw power that a NSWR makes available, consider this.

The exhaust jet makes a very effective directed energy particle beam weapon.

Without cooling, no physical substance we can make is capable of standing up to the conditions in the reaction chamber, the exhaust nozzle, or for several hundred thousand kilometers in whatever direction the exhaust is being directed.

In other words, the main engine of a NSWR ship is a very very potent main gun, with a pretty good range.

Even with cooling, the reaction chamber and exhaust nozzle of a NSWR could get so hot that if seen from 1LY distance they would appear brighter than most stars to the naked eye.

I see 3 big problems with NSWR propulsion, but I also see potential solutions for each one.

The biggest problem I see is the reaction chamber and exhaust nozzle. Basically, no known material can stand up to the stresses created by normal operation of a NSWR. There's actually a pretty easy solution for this. Don't use a physical reaction chamber or exhaust nozzle. Use magnetic fields instead. The material challenges will still be nearly insurmountable, but "almost impossible" is better than "clearly impossible".

Next, how do you shield the payload? Best solutions that I can come up with:

Tungsten "shadow" shield closest to the reaction (gamma ray shielding)

Nuclear salt water (fuel) followed by regular water (propellant) tanks used as intermediate shielding. (neutron shield)

I think that a NSWR would actually carry around more "regular water" than "nuclear salt water", even if it didn't need the regular water for radiation shielding. The regular water would be injected around the reaction chamber's walls, and the nuclear salt water would be injected into the center. The reaction from the nuclear salt water heats up the regular water, and both are exhausted out the back. This improves thrust at the cost of specific impulse. NSWR can get up to 482,140s ISP when using just nuclear salt water, so even if a lot of that is traded for thrust you still end up with a rocket that's hundreds of times more efficient than anything we have built so far.

Finally, another magnetic field surrounding the payload (alpha/beta/charged particle shielding)

Finally, how do you keep the fuel from going boom before it gets to the engine? This one seems almost too easy, I must be missing something. Here goes anyways:

The best way to prevent the nuclear salt water from reacting outside the engine is to have as little of it on the ship as possible at any one time.

The engine's operation creates a convenient source of neutron radiation that can be put to use to breed nuclear fuel from materials that can be used as radiation shielding. Store as much water salted with uranium-238 (depleted uranium) as you want, but pipe it thru channels around the reactor to expose it to the large neutron flux before it's used to transmute it to the fissile isotope that is actually used by the engine. This might need a buffer tank to wait for the irradiated fuel to transmute into the useful isotope, but that should still be a smaller tank than carrying the salt water of the useful isotope directly.

I call this a Nuclear Breeder Salt Water Rocket, because it uses the currently reacting fuel to breed more fuel for itself.

In simpler terms, the solution is to carry large amounts of "relatively inert stuff that can be turned into highly volatile fuel", instead of carrying "highly volatile fuel".

[jfull]

Yup, the only thing you need to do to make a basic nuclear bomb explode is to rapidly assemble a supercritcal mass of weapons-grade fissile material. The natural decay, creating neutrons, will get the reaction started, and since its supercritical, it will rapidly go out of control. That's IT.

It can be challenging assembling the supercritical mass quickly enough, however; if you do it too slowly, the very start of the runaway reaction will blow the components apart, and the weapon will fizzle as a very low yield dud. So the key is to assemble a supercritical mass very quickly, before it can blow itself apart. High explosives are necessary in all the designs that I'm aware of, but the designs are still very simple.

This is why Iran is such a worry. Having the ability to produce highly enriched uranium makes you a defacto nuclear weapons state.

This just in:

government agents have placed Velocity on the international no-fly list

[megatiger78]

how do I quantum vascular then? that sounds less complex.............

[meve12]

This rocket would not use nuclear explosions at all.

Seeing as the Zubrin drive uses what amounts to a continuous prompt super-critical reaction contained in a nozzle that amounts to little more than half a bowl with fuel valves, that's somewhat academic.

Although, unlike the majority of fission power plant designs, breaching the tank would indeed result in an explosion.

[Kibble]

Shame this conversation has devolved into pointless semantics!

[11of10]

@SciMan I like the idea of enriching your fuel just before inserting it into the "combustion chamber" but, that's all well and good once you have the thing running. How do you start it in the first place? Even if you really can enrich it enough with your idea for it to have merit, and I don't know the answer to that.

[SciMan]

Right... the breeding ratio. As I said in my earlier post, I was almost certain I was missing some detail that makes my idea not work well, if at all.

It "should" still work, but it might need to carry nuclear reprocessing gear as well as all the other stuff. With all the thrust and specific impulse a NSWR is capable of, it shouldn't really make all that much difference to the performance.

I thought about this idea for a while, so I have the start-up problem solved too. Basically, there's a tank of "activated" nuclear salt water that is used for start up, and it's refilled by tapping off the supply going to the engine after start-up.

To explain it properly, here's a propellant flow diagram I made to help myself keep track of what's going on in the engine.

Nuclear salt water flow:

(Large U-238 salt water tank)-->(Reaction chamber cooling and fuel breeding passages)-->(Small Pu-239 salt water tank)-->(injection into reaction chamber)

Starting is the same as for a normal NSWR, with the starting fuel taken from the Pu-239 salt water tank.

Once the engine is running, that tank is refilled by the engine's normal process of breeding U-238 salt water into Pu-239 salt water.

As described in my previous post, regular water can also be injected just after the reaction chamber. This increases total propellant mass flow rate, which increases thrust at the cost of reducing specific impulse. That's all well and good, but it also means that you end up using a lot less nuclear salt water for the same thrust level, which means using less Uranium or Plutonium. Uranium and Plutonium at the enrichment levels required aren't exactly cheap, so the less of it you can use the better.

For that reason, I expect that at least the first few non-breeding NSWR's built will use the "nuclear salt water + extra water" mode almost exclusively, in order to save nuclear fuel.

If that wasn't enough reason to inject extra water, the specific impulse of a NSWR operating without additional water (482,140s) is sub-optimal for the great majority of potential applications.

For interplanetary travel, it's massively overkill, but for interstellar travel, it's insufficient by several orders of magnitude.

The same amount of power applied to a much higher flow rate of reaction mass would still be "good enough" for interplanetary travel, and it would produce quite a lot more thrust.

Basically, the only drawbacks to injecting additional water into the exhaust for more thrust are a slight increase in the complexity of the engine's plumbing, along with the mass of the additional water. In other words it's practically free, and who says no to free thrust?

Edited December 10, 2014 by SciMan

[NERVAfan]

If that wasn't enough reason to inject extra water, the specific impulse of a NSWR operating without additional water (482,140s) is sub-optimal for the great majority of potential applications.

For interplanetary travel, it's massively overkill, but for interstellar travel, it's insufficient by several orders of magnitude.

You can't really do "several orders of magnitude" better than that, as that's an exhaust velocity of a bit over 1.5% of the speed of light!

If you can manage a mass ratio of 10, you could get a cruise speed of almost 5500 km/s (about 1.8% of lightspeed) which would get you to Proxima Centauri in 236 years, Barnard's Star in 327 years, or Epsilon Eridani in 582 years. That's quite workable for a generation ship IMO.