[amagmasia]

A Preliminary Engineering Design for a Nuclear Photonic Spacecraft Engine Operating at Ultra-high Temperatures.

By ultra-high temperatures I don't mean the highest temperature of a currently-operating nuclear reactor, about 825 K. I don't even mean the temperature of a proposed "Very High Temperature Reactor", 1270 K. I mean a temperature of 2200 K, well above the melting point of uranium.

Insane, possibly, but I think it's borderline technically feasible with current technology, provided the reactor is in a microgravity environment and provided the temperature remains constant.

A photonic rocket engine is one powered by the momentum of the photons it emits. The radiation increases as the fourth power of the temperature so we want it as hot as possible - this one is powered directly off the heat of the nuclear reactor. The upper limit on temperature of 2200 K is given by the boiling point of beryllium moderator at low (in this case 15% atmosphere) pressure. The low pressure helps with structural integrity. If the operating temperature is dropped much below 2200 K, dissolved uranium has a tendency to separate into a separate liquid phase, which would be very bad for corrosion of the reactor cavity.

Figure 1. Expanded view of the ultra-high temperature nuclear photonic spacecraft engine. The three layer heat shield at top is adjacent to the carbon moderator layer, which seals against the hotplate-impeller assembly to create the reactor cavity. The rim drive, shown lower left, is mounted off the carbon moderator layer.

The reactor core is contained in metal tungsten-184 obtained by chemical vapour deposition from gas centrifuge. The nuclear fuel is plutonium-239. The lowest energy fission resonance of 239Pu at 0.3 eV plays a big role in minimising the reactor weight (by minimising the amount of moderator required) while at the same time providing reactor stability (by ensuring that the fission cross section decreases with increasing temperature). Some natural uranium can be mixed with the plutonium if desired for use in breeding more 239Pu internally.

There is a heat shield made from tungsten-184, a mix of molybdenum isotopes 92Mo and 94Mo, graphite and aluminium. The molybdenum isotope mix comes from a single pass of natural molybdenum through a gas centrifuge. This heat shield was very easy to design and only needs to be three layers thick.

The hotplate and impeller, at the bottom of Figure 1, is rotated by the rim drive, and this forced-convective action keeps the reactor temperature very close to constant. Wall material 184W is pre-dissolved in the molten beryllium moderator to form a saturated solution. The use of this saturated solution ensures that the rates of erosion and deposition of wall material match, minimising long-term damage to the walls.

An alternative use for this reactor design would be as an optical pump for a high-powered laser, as shown in Figure 2.

Figure 2. Adaption of ultra-high temperature nuclear reactor design as an optical pump for a laser power source. As shown it has a length of about 15 metres and diameter of about 2 metres. The end cap and rim drive are not shown. Inside is the laser. Around that is the reactor cavity with 8 rotating vanes. Around that is the carbon moderator and on the outside is the three layer heat shield.

For full details see http://freepages.misc.rootsweb.ances..._spaceship.pdf

Comments?

[ftpsoft]

It's imaginative...It's innovative...And as long as some of us are thinking and talking of methodology of getting to the stars, we'll eventually work something out.
Other then that molly, I'm not really qualified to comment, other then to say, that for every physicist/Engineer that is able to think outside the square and say yes it worth a try, you will probably have a dozen who will rubbish the idea.
That is human nature with many examples throughout history.

But thankfully we do have orginizations throughout the world, with little financial backing, and probably thought of as "nutters" by some, that are thinking of projects to take us to the stars all the time.


Projects such as "Daedalus" and "Icarus" and even "Build the Enterprise" Dan concept are examples.


"A journey of a thousand light-years begins with a single step."
Confucius:


It could be said that "The best way to predict the future is to create it."

[Dokescoonse]

Insane, possibly, but I think it's borderline technically feasible with current technology, provided the reactor is in a microgravity environment and provided the temperature remains constant. If the drive was at producing thrust, wouldn't that mean that the reactor was no longer in a microgravity environment?

Success would lead to failure?

[DownloadADOBEsoftware]

Interesting ...

Some questions.
How do you regulate the power?
How do you start & stop it?
How do the photons escape from the bottom?
Roughly how much would it mass?
Roughly how much thrust would it make and for how long?
Would you get a better impulse by using all that (heat) energy from the same type of reactor to expel a gas or other medium through a nozzle?
Does the heat shield get radioactive eventually?

Apologies if those are answered in your paper.

[maks_holi]

How do you start & stop it? With a button, of course. Put a button on anything and you can magically turn it on or off. It's a fundamental law of physics

[CuittisIL]

What colours does it come in?

[voksveta]

non-spectral purple

[amagmasia]

Thanks for the questions !!

> If the drive was at producing thrust, wouldn't that mean that the reactor was no longer in a microgravity environment?

Did check that. Am expecting an acceleration of the rough order of 1.7e-5 m s-2 which still counts as microgravity. At that acceleration it doesn't get to the nearest stars very fast, but significantly faster than normal rocket engines. The danger, though, is that the impeller rotation will create significant gee forces, which I haven't calculated yet.

> How do you regulate the power?

Straight to the heart of a major difficulty. Of the several options, the only two I can see with a chance of working are first to slowly pump in more plutonium (it has a low melting point so this is not difficult) and second to slowly remove a high temperature neutron absorber such as natural tungsten through slots in the carbon moderator layer. Possibly both.

> How do you start & stop it?

To stop it, first let it die. When it gets below a critical temperature create a gap between the reactor cavity and the carbon moderator to cool it more quickly.

The beryllium will freeze releasing as liquid the fission products lanthanum to promethium as well as waste plutonium (Pu-239 and Pu-240) which are removed. Other products (Cs, Ba, Am) will have already been removed as gases while the engine is in operation.

Then slot a new engine into the gap between the old one and the carbon moderator layer and use the heat from this new engine to re-liquefy the beryllium for pumping out into a third engine to be used later. The third engine will already (before lift-off) be pre-filled with the elements needed to combine with the incoming beryllium to make a working reactor.

> How do the photons escape from the bottom?

The tungsten hotplate at the aft end of the engine will be of the rough order of 10 mm thick. Photons are generated from the tungsten hotplate in very much the same way as from the tungsten filament in an incandescent light bulb, direct radiation.

> Roughly how much would it mass?

The height and diameter of the reactor disk are still not settled. The diameter is essentially a free variable, for more overall thrust you simply make it bigger, and the engine mass increases in direct proportion to the thrust. My initial design had a hotplate surface area of 100 m2, which is a diameter of 11.3 m, for no particular reason. This could be increased up to for instance 30 m if required.

Similarly the height of the reactor disk is still not settled. By adjusting the number of fissions per second (which is a non-trivial calculation) the height could also be varied within limits, with shallower engines burning out faster but weighing less so being replaced more often.

That said, it is possible to get a very accurate estimate of the total pass of all engines per square metre of hotplate area as a function of operation time. The reactor generates 1.4 MW/m2. The proportion of plutonium atoms fissioned may be as high as 30%. Using this, 1 kg of plutonium generates 2.5*107 Megajoules. So the plutonium usage is 1.78 kg m-2 y-1. Uranium is included with the plutonium here but discounted by a factor of 2.

By atomic number within the reactor fluid (plus casing) we have Pu:Be:W of 1:62:7, by mass that is 1:2.3:5.1
(Unfortunately for me, a new calculation today of the solubility of tungsten in beryllium more than doubles the total tungsten needed). However, the beryllium and tungsten are largely recycled, set the recycling efficiency to 90%. That gives a mass ratio Pu:Be:W = 1:0.23:0.51. So the engine mass is 3.1 kg m-2 y-1. The masses of carbon, molybdenum and aluminium are small enough to ignore at this level of accuracy. The uranium is recycled.

For a thousand year journey that's an engine mass of 3.1 ton m2, and over a 100 m2 hotplate that becomes 310 tons or over a 30 m diameter hotplate 2,200 tons.

The mass will decrease as spent engines are jettisoned, as in a multi-stage rocket.

> Roughly how much thrust would it make and for how long?

0.00445 N m-2 thrust, so with a 100 m2 hotplate that becomes 0.4 N or over a 30 m diameter hotplate 3 N. Not much, but slightly better than wikipedia's estimate of "10-5 to 1" Newtons.

I don't know how long an individual engine will last before tuning in a spare, the calculations are a bit complicated. Overall using every spare engine, as long as it takes, eg. 1000 years.

I have not calculated the thrust from the optically pumped laser option. Wikipedia has this to say about it:
Photonic Laser Thruster (PLT) is a pure photon laser thruster that amplifies photon radiation pressure by orders of magnitude by exploiting an active resonant optical cavity formed between two mirrors on nearby paired spacecraft. PLT is predicted to be able to provide the thrust to power ratio (a measure of how efficient a thruster is in terms of converting power to thrust) approaching that of conventional thrusters, such as laser ablation thrusters and electrical thrusters. In December 2006, Dr. Young K. Bae successfully demonstrated the photon thrust amplification in PLT for the first time with an amplification factor of 3,000 > Would you get a better impulse by using all that (heat) energy from the same type of reactor to expel a gas or other medium through a nozzle?

Excellent question. Because the exhaust speed is much lower for a thermal nuclear powered system, the thrust is much better. But on the other hand for the photonic nuclear powered system there's no need to carry the very large reaction mass needed by the thermal system. I'd very much like to see someone do a direct comparison.

> Does the heat shield get radioactive eventually?

I initially thought the answer to this was "yes", but realise now that that isn't necessarily the case. I'll check. The carbon moderator between the reactor and the heat shield blocks most of the neutrons and all the alpha and beta rays but not all of the gamma rays. The components of the heat shield, 184W, C, the two molybdenum isotopes are selected for minimal neutron damage.

I'll explain why the answer isn't necessarily "yes". 184W can sometimes absorb at least 26 neutrons without becoming radioactive for more than a couple of days, and some of the intermediates have extremely low neutron absorption cross sections, while others would evaporate off into space before absorbing any more neutrons.

> Apologies if those are answered in your paper.

None were. Superb questions, you can ask me questions any time.

Correction to paper

A saturated solution of tungsten in beryllium at 2200 K is 10% tungsten by atomic number, not 4% which was my previous best estimate.

An alternative to natural beryllium as a liquid moderator is the isotope boron-11. This is 80% of natural boron so would have to be isolated out. Boron-11 has lower neutron absorption cross sections than beryllium and a much higher melting point allowing 2400 K operation at a much lower pressure than possible with beryllium. A downside is that tungsten is very soluble in molten boron; at 2400 K a saturated solution is roughly 7% by atomic number (perhaps a little less), but it only takes a temperature difference of 10 degrees to change the saturated solution by 1%.

[amagmasia]

> What colours does it come in?

Near infra-red, with a slight red glow.

[amagmasia]

>> Would you get a better impulse by using all that (heat) energy from the same type of reactor to expel a gas or other medium through a nozzle?

> I'd very much like to see someone do a direct comparison.

I tried a direct comparison (as well as I could) and concluded that for the same top speed a thermal nuclear engine that expels gas or similar would have to be at least 12.5 times as heavy as the photon nuclear reactor design here. A thermal nuclear engine would have a higher thrust so would accelerate to top speed and then cruise but even so would need to be at least 6 times heavier than a nuclear photonic engine to achieve the same travel time. And weight matters.