Author Topic: VASIMR as Plasma core fission rocket  (Read 2868 times)

Offline Elmar Moelzer

  • Senior Member
  • *****
  • Posts: 2858
  • Liked: 464
  • Likes Given: 710
Re: VASIMR as Plasma core fission rocket
« Reply #20 on: 04/21/2017 08:26 AM »
But won't that just effectively make it a nuclear thermal rocket (if a rather contrived one)? Wouldn't the 'additional thrust' from exhausting coolant hydrogen make up most of the thrust.
I guess it depends on how much hydrogen you have to bleed through to cool the rocket. If it is very little, it will probably not matter much.

Offline Rei

  • Full Member
  • ****
  • Posts: 467
  • Iceland
  • Liked: 179
  • Likes Given: 58
Re: VASIMR as Plasma core fission rocket
« Reply #21 on: 04/21/2017 03:10 PM »
But won't that just effectively make it a nuclear thermal rocket (if a rather contrived one)? Wouldn't the 'additional thrust' from exhausting coolant hydrogen make up most of the thrust.
I guess it depends on how much hydrogen you have to bleed through to cool the rocket. If it is very little, it will probably not matter much.

What's the point of the plasma then, if you're just exhausting hydrogen at ~800-900 sec ISP? Why not a solid core and save yourself all of the containment issues, simplify moderation, get a far better mass fraction, and so on? 

Regardless of how you try to arrange things, if you want the hydrogen to carry away the fission energy, it has to vastly outmass the fissile fuel.  All the moreso if it has to do so at temperatures that won't destroy cooling channels.
« Last Edit: 04/21/2017 03:16 PM by Rei »

Offline as58

  • Full Member
  • ****
  • Posts: 668
  • Liked: 208
  • Likes Given: 144
Re: VASIMR as Plasma core fission rocket
« Reply #22 on: 04/21/2017 06:37 PM »
Obviously what is needed is aneutronic fission. ;) Preferably also without gamma radiation.

Offline Elmar Moelzer

  • Senior Member
  • *****
  • Posts: 2858
  • Liked: 464
  • Likes Given: 710
Re: VASIMR as Plasma core fission rocket
« Reply #23 on: 04/21/2017 11:16 PM »
Regardless of how you try to arrange things, if you want the hydrogen to carry away the fission energy, it has to vastly outmass the fissile fuel.  All the moreso if it has to do so at temperatures that won't destroy cooling channels.
I am not so sure about that. IIRC most of the VASIMIR rockets are bleeding small amounts of hydrogen through the cooling channels after the rocket engine has been turned off to keep the engine cool enough.
Those were quite obviously only very small amounts (or you could have just kept firing the engine).
Obviously it depends on how hot the walls of Robot's rocket engine will get. It is not my design but his, so he would have to be the one to answer that question.

Offline Rei

  • Full Member
  • ****
  • Posts: 467
  • Iceland
  • Liked: 179
  • Likes Given: 58
Re: VASIMR as Plasma core fission rocket
« Reply #24 on: 04/22/2017 08:07 AM »
Regardless of how you try to arrange things, if you want the hydrogen to carry away the fission energy, it has to vastly outmass the fissile fuel.  All the moreso if it has to do so at temperatures that won't destroy cooling channels.
I am not so sure about that. IIRC most of the VASIMIR rockets are bleeding small amounts of hydrogen through the cooling channels after the rocket engine has been turned off to keep the engine cool enough.

Which has nothing to do with the ratio of nuclear fuel to hydrogen required to absorb the fission energy in a nuclear-equivalent of VASIMR.

The concept, unless I misunderstood Robot, is that a fission reaction is what's making the plasma. Which means that you have a tremendous neutron flux in your moderator-reflector, corresponding to the level of fission needed to self-heat to a plasma (it's even worse if you want the nuclear fuel to be solid and only the hydrogen plasma). Now, pick from from the following scenarios:

1) Hydrogen fully cools the moderator before entering the plasma: then the rate of exhausting the fuel-hydrogen mixture must match; the nuclear fuel is only using the most utterly miniscule fraction of its potential energy on average before being exhausted, to the point where you might as well just be burning the hydrogen instead.

2) The fuel is being burned up: then pick one of the two:

2a) The hydrogen is grossly insufficient by many orders of magnitude to handle the heat load

2b) The hydrogen is outmassing the fuel in the core by many orders of magnitude, and thus shutting off fission.

There's no working around it. No matter your geometry, a large chunk of the fission energy is going into the walls. If you're cooling the walls and exhausting hydrogen, then you're facing the same limits as NERVA in terms of performance. Wherein, why the heavy, convoluted design?  "Cooling" is something that happens on the order of a fraction of an eV-per-atom**.  Fission is something that happens on the hundreds-of-MeV-per-atom scale. And a good chunk of that hundreds-of-MeV-per-atom ends up in the walls. Unavoidably. You simply cannot cool it down non-radiatively without either grossly outmassing the fuel by many orders of magnitude, or by only using the tiniest fraction of the fuel before exhausting it. And both render the exercise pointless.

** Hydrogen's specific heat is ~14,3 kJ/kg, and for it to be a "cooling" fluid it must be colder than the object it's cooling. 1 eV per hydrogen atom = 96,5 MJ/kg hydrogen.  Uranium fission ~= 200MeV per atom . Yes, the fissile fuel is 2 orders of magnitude heavier per atom than hydrogen, but that does little to close the gap.

If you want high thrust from a fission rocket, you want nuclear thermal.
If you want high impulse from a fission rocket, you want a fission fragment rocket.
« Last Edit: 04/22/2017 08:22 AM by Rei »

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 26074
  • Minnesota
  • Liked: 6087
  • Likes Given: 4498
Re: VASIMR as Plasma core fission rocket
« Reply #25 on: 04/22/2017 06:21 PM »
Good points. Might need the hydrogen itself to act as the neutron reflector, or simply have a big enough core that no reflector is needed. You still have to cool the coils, but the neutrons could mostly just fly out into space. Additional hydrogen could be added to increase thrust, but not so much that the plasma state is unable to be maintained.

These requirements may mean the engine would have to be HUGE.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Rei

  • Full Member
  • ****
  • Posts: 467
  • Iceland
  • Liked: 179
  • Likes Given: 58
Re: VASIMR as Plasma core fission rocket
« Reply #26 on: 04/22/2017 07:32 PM »
Good points. Might need the hydrogen itself to act as the neutron reflector, or simply have a big enough core that no reflector is needed. You still have to cool the coils, but the neutrons could mostly just fly out into space. Additional hydrogen could be added to increase thrust, but not so much that the plasma state is unable to be maintained.

These requirements may mean the engine would have to be HUGE.

The neutrons don't just fly out into space.  By definition of a moderator, it ends up capturing the majority of the energy of the neutrons. A moderator thermalizes neutrons - brings them down to the roughly the same energy  as the moderator itself by repeated interactions, primarily elastic. Which is virtually all of the energy of the neutrons. Some will ultimately be absorbed and some will ultimately escape, but to maintain slow fission, the vast majority of the neutron energy must go into the moderator, regardless of what ultimately happens to them.

The whole point of moderating down is about cross sections. If you're not familiar with the concept, picture that you're throwing tennis balls randomly down a hall in which someone has hung up plates at random positions. The odds that you'll end up hitting a plate is relative to the cross section of the interaction between your tennis ball and the plates. If you increase the size of the ball or increase the size of the plates, the odds of a collision rises - and vice versa.

With nuclear interactions, it's not exactly the physical size that matters, but it makes for a nice analogy, and hence cross sections, with the units (barns) based to be in the ballpark of typical physical cross sections.

Anyway, elastic scattering, (n, gamma) and fission cross sections typically look like these:

http://www.nndc.bnl.gov/sigma/getPlot.jsp?evalid=15321&mf=3&mt=2&nsub=10
http://www.nndc.bnl.gov/sigma/getPlot.jsp?evalid=15321&mf=3&mt=102&nsub=10
http://www.nndc.bnl.gov/sigma/getMF5.jsp?evalid=15321&mf=5&mt=18&nsub=10

You'll notice how cross sections tend to increase by many orders of magnitude once you get down into the thermal spectrum, which means that it becomes far easier to use neutrons for whatever purpose (including fission). 

This is getting a bit off topic, but the short of it is, you give up the overwhelming majority of the neutron energy into the moderator (excepting in fast reactors, but for a wide variety of reasons they're generally not proposed for space usage), even when you're losing neutrons. And if you're doing any simple exhaustion of hydrogen at all to get rid of heat then you might as well go fully nuclear thermal, because your ISP is going to be poor due to that loss (aka, simultaneously combining a 800 ISP thruster with a 10000 ISP thruster gives you something far more like an 800 ISP thruster). 

Now, it seems you're trying to work around this by scaling up to the point that you don't need an externally located moderator, that it's located in the plasma? Scaling up actually dumps more neutron radiation into your magnets, but the ratio of thrust to magnet heating will rise with scale. But how hot are you having the plasma? ISP is relative to your plasma temperature. But the thermal radiation from the plasma is proportional to its temperature to the *fourth* power. As as mentioned previously, you can't have some sort of near-perfect reflectivity, because even if it starts out almost perfectly reflective, the intense thermal and neutron fluxes damage your mirroring over time.  So the fission isn't actually upping the maximum ISP of your VASIMIR thruster, nor its max thrust; it's just replacing the means of heating it, at the cost of having a more hazardous fuel with a higher atomic mass, irradiating your spacecraft, and in particular your superconducting magnets.  And remember that it's *very* thermodynamically expensive to remove heat from extremely cold objects. In VASIMR they only have to deal with reflecting radiated heat from the plasma, but in your case they're being heated internally.

There are of course some theorized designs between simple nuclear thermal and a purely plasma driven design. There's one gas/plasma core concept (open gas cycle NTR) where they transfer the energy to hydrogen while trying to minimize the mixing of hydrogen and fuel by fluid dynamics processes to decrease the rate at which fissile fuel is exhausted, and to allow the hydrogen to be ejected rapidly for maximum thrust and minimum engine heating. But an effective means to keep the two fluids separate is still a topic of research. There's also a closed gas cycle NTR version which I find kind of interesting, sometimes called a "nuclear lightbulb" - the fission plasma is inside a fused quartz / fused silica tube (or other high temperature transparent material), preventing exchange / leakage between fluids. The thermal energy of the radiating fissile fuel passes through the transparent tube; some is absorbed in the process, but the concept is to keep the tube cool enough that it doesn't melt.  Again, very immature, but still, an interesting idea.  The biggest problem is neutron radiation blackening, which is particularly a problem at short wavelengths.

I once played around with the concept of a multimode fission rocket which was to function as a nuclear lightbulb while in the atmosphere but a fission fragment rocket in space. Kind of fun project. But it's a big materials challenge. Radiating from a dusty or misty core means longer wavelength light, which means that the blackening isn't as much of a problem, but it also means you need a lot more radiating surface area.
« Last Edit: 04/22/2017 07:42 PM by Rei »

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 26074
  • Minnesota
  • Liked: 6087
  • Likes Given: 4498
Re: VASIMR as Plasma core fission rocket
« Reply #27 on: 04/22/2017 08:31 PM »
I wasn't thinking of moderating the neutrons. I was thinking a fast neutron reactor.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Stormbringer

  • Full Member
  • ****
  • Posts: 1266
  • Liked: 218
  • Likes Given: 80
Re: VASIMR as Plasma core fission rocket
« Reply #28 on: 04/22/2017 09:17 PM »
There is an article today about Chinese researchers having developed a neutron source that is 100 fold better than current state of the art and they believe they can make it a one thousand fold increase:

https://phys.org/news/2017-04-laser-technique-neutron-yield.html
When antigravity is outlawed only outlaws will have antigravity.

Offline Rei

  • Full Member
  • ****
  • Posts: 467
  • Iceland
  • Liked: 179
  • Likes Given: 58
Re: VASIMR as Plasma core fission rocket
« Reply #29 on: 04/23/2017 02:23 PM »
There is an article today about Chinese researchers having developed a neutron source that is 100 fold better than current state of the art and they believe they can make it a one thousand fold increase:

https://phys.org/news/2017-04-laser-technique-neutron-yield.html

For some reason the actual paper isn't loading for me, so I can't see exactly what's being mentioned.  100 fold "better" in what context, and state of the art "what"?  I can see a graph which shows 1e8-1e10 neutrons per pulse, and no info on the pulse rate (but since it involves ICF on hohlraums, it's going to be low). Nuclear reactors and spallation neutron sources produce many orders of magnitude higher neutron fluxes than that per second.  It might be a good yield for "fusion neutrons", which are single-energy and hotter than fission "evaporation" neutrons (although colder than spallation neutrons before multiplication).

Offline Rei

  • Full Member
  • ****
  • Posts: 467
  • Iceland
  • Liked: 179
  • Likes Given: 58
Re: VASIMR as Plasma core fission rocket
« Reply #30 on: 04/23/2017 02:59 PM »
I wasn't thinking of moderating the neutrons. I was thinking a fast neutron reactor.

Fast reactors are interesting, although they have disadvantages that have generally led to avoiding them for space missions.  They're sort of balanced on a knife's edge, more wanting to either runaway or shutdown entirely (less Doppler broadening) in comparison to a thermal reactor. The neutron spectrum hitting your structural materials is hard, rather than moderated, and thus more difficult to shield / more penetrating (although fast reactor cores are more compact, so the surface area is less). It's usually  harder to ensure longevity with a fast reactor. Proliferation is more of a concern, since you have to enrich them more because cross sections are so much lower at higher energies. And experience with them is lower than with thermal reactors. But don't get me wrong, they are still interesting. Small volume, high burnup, fast responsiveness, etc.

But they don't get you around the fact that if you have a plasma in the center of a core - regardless of the neutron flux - its temperature is going to be limited by your ability to deal with the thermal radiation it gives off.  Nuclear or not. Which limits both thrust and ISP. Fission fragment reactors work around this by having the fuel itself not actually be that hot; it's giving off fission fragments, which are *extremely* high energy, but as ions they're constrained by magnetic fields and immediately leave the core, while the heavy grains / droplets have a far higher ratio of mass to charge and are constrained by a Penning trap.  This doesn't work around thrust limitations, mind you - hence my playing around with dual-mode reactors ("nuclear lightbulb" NTR for in atmosphere, fission fragment in space)

Tags: