Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Thorium Propulsion

  1. Jan 5, 2007 #1
    There's been another small spate of news articles in the past couple of weeks about Thorium, due to its safer and more environmentally friendly characteristics in comparison to Uranium.

    I'm wondering if there has been any fresh progress or advancements in the developmental research of Thorium as a nuclear fuel -- particularly Accelerator Driven sub-critical reaction systems.

    I'd always wondered if Thorium's potential for use as a controlled sub-critical nuclear power source might one day make it suitable for space propulsion. Perhaps a Thorium nuclear power source could be used to power a VASIMR or other type of rocket propulsion system, enabling a single-stage spacecraft to travel directly from the Earth to the Moon.

    I was also prompted by the latest news coverage of Blue Origin's private spacecraft test launch, which once again reminds of the challenges faced in using the limited energy in chemical fuels to attain orbit. If only we had much higher energy density fuels to work with, then there would be more margin for flexibility in the design of aerospace craft. (Not just spacecraft, but even large high-speed heavy-lift aerial intercontinental cargo transports to service our global economy.)

    I think humongous nuclear-powered launch vehicles would be great, but those old 1950s designs like Orion and NERVA are pretty scary, and no reasonable person would consider them safe.

    If Thorium could be harnessed sub-critically to power a rocket propulsion system, then what would the limit be on launch vehicle size and payload lifting capacity? I'm imagining you could build a launch vehicle the size of a Supertanker or the Empire State Building, if you wanted to.

    Last edited: Jan 5, 2007
  2. jcsd
  3. Jan 5, 2007 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    This thread belongs in nuclear engineering.

    There seems to be a lot of confusion in the media then. Thorium fuel needs a fissile component, e.g. U-233 or U-235, and then the Th-232, which is fertile, can be used to produce more U-233. Even with a thorium fuel cycle, one cannot avoid 'fission products'. It does have some advantages like higher thermal conductivity than UO2, but it still has many of the disadvantages.

    IIRC, India is using thorium fuel, but I am not sure the extent. Russia has proposed a thorium fuel cycle for its VVERs.

    Thorium can be mined from monazite sands (http://en.wikipedia.org/wiki/Monazite) which are fairly abundant.

    http://www.ead.anl.gov/pub/doc/thorium.pdf [Broken]

    http://www.world-nuclear.org/info/inf62.htm [Broken]

    http://www.princeton.edu/~globsec/publications/pdf/6_3galperin.pdf [Broken]


    As for propulsion - there are numerous other issues involved - but ultimately it comes down to specific power and thrust.
    Last edited by a moderator: May 2, 2017
  4. Jan 5, 2007 #3
    Astronuc, no, the method India is pursuing is similar to that pioneered by the US and also Germany, who had previously built Thorium reactors which use U-235 or Plutonium as a neutron spallation source. However there is the method proposed by Italian Nobel laureate Carlos Rubbia, for a sub-critical Accelerator Driven Thorium reactor which uses a proton beam for spallation. But it's not clear if the neutron economics sufficiently support this. It's still being researched in the lab.
  5. Jan 6, 2007 #4


    User Avatar
    Staff Emeritus
    Science Advisor

    Yes, I am aware of that. By thorium fuel, I was referring to a fuel with a primarily thorium matrix with U-235 or other fissile nuclide added, because Th-232 has a very low thermal fission cross-section. Th-232, like U-238, is fertile. Whereas U-238 absorbs a neutron then through 2 successive beta decays is converted to Pu-239, Th-232 is converted to U-233, which is fissile and can be used instead of U-235, which must be obtained from an enrichment process. The necessary enrichment is reduced if heavy water is used.

    The VVER and CANDU reactors/fuel designs are better designs for the thorium cycle.

    http://www.world-nuclear.org/info/inf80.htm [Broken]



    http://www3.inspi.ufl.edu/icapp07/program/abstracts/7367.pdf [Broken] - Abstract
    Last edited by a moderator: May 2, 2017
  6. Jan 6, 2007 #5

    Andrew Mason

    User Avatar
    Science Advisor
    Homework Helper

    Your post reminds me of a cartoon from the 70s showing a Soviet bureaucrat showing off his new multifunctional electronic wrist watch but lamenting the fact that he has to carry around this huge anvil-like battery. Accelerators for anything other than electrons tend to be rather large and heavy!!!

  7. Jan 6, 2007 #6


    User Avatar
    Staff Emeritus
    Science Advisor

    And inefficient, because of the large energy input.
  8. Jan 6, 2007 #7

    Andrew Mason

    User Avatar
    Science Advisor
    Homework Helper

    I am no expert on this by any means but it appears that the advantages of thorium are that:

    1. all of the thorium fuel can be consumed in a slow reactor. With uranium fuel, less than 1% can be used,

    2. since U233 produces more neutrons per fission, the reactor can continue to operate with higher concentrations of fission products in the fuel so the fuel lasts longer in the reactor.

    I don't see any advantage to using thorium in a rocket ship, other than the savings in not having to cart around the extra U238. But that could be avoided simply by using highly enriched uranium.

  9. Jan 6, 2007 #8
    Well, my post was motivated by the fact that chemical fuel seems to be so low in energy density, as to offer very little margin or flexibility in launchcraft design.

    So my thought then went to nuclear fuel, looking to discuss any possible useful candidates for this purpose.

    As for accelerators being bulky, there are laser-wakefield accelerators which are being heralded as a new generation of tabletop accelerators, able to achieve astonishing acceleration gradients in a small dimension size. Obviously, these currently work for electrons only, but it makes me wonder whether the technology couldn't be adapted for protons, by using a higher (gamma?) frequency? So perhaps you could have a gamma-ray laser firing inside some pressurized hydrogen gas, to produce a wakefield beam of protons. Is that possible?





    Last edited by a moderator: Apr 22, 2017
  10. Jan 7, 2007 #9

    Andrew Mason

    User Avatar
    Science Advisor
    Homework Helper

    Of course, for a rocket one has to eject mass to provide propulsion. If you could get electrons moving with speed sufficiently close to c, they might be able to carry significant of momentum. The higher the gamma, the fewer electrons you would need to eject per unit time.

    [tex]F = dp/dt = \gamma m_ec\frac{dn}{dt}[/tex] where dn/dt = the number of electrons per unit time being ejected.

    But even if you could make such a small powerful accelerator you would have an increasing problem with electron ejection. As you ejected electrons, the charge of the rocket ship would become more positive. One mole of electrons (which is about half a milligram), represents a charge of about 37,000 Coulombs which is an enormous charge. So your accelerator would be working much harder to accelerate those electrons.

  11. Jan 7, 2007 #10


    User Avatar
    Staff Emeritus
    Science Advisor

    When one looks at the atomic/molecular level, one finds nuclear with MeV per reaction vs eV/reaction for chemical - BUT - when one looks at the reactions/unit mass or /unit volume, the difference is not that great. There are physical (= engineering) constraints, e.g. melting temperature, creep rate, yield/tensile stress, stress to rupture, which all lead to failure of a system.

    And what is the power supply? This is the issue with proplusion - the power supply/plant and energy source (fuel) must come along for the ride.

    Any electrically charged propellant must ultimately be neutralized since the electrons will eventually try to recombine with the + charges, and that reversal of momentum would oppose the forward momentum of a spacecraft.

    As for 'table top' systems, what is the power level? The power supply? Can they be scaled?
    These are basic design considerations.

    Also regarding VASIMR - so far it's not viable.

    The ideal propulsion system produces momentum in one direction - which is approximately what happens with an electrostatic system. The key problem for most propulsion systems is that a hot propellant produces 'pressure' in the lateral direction and so one needs an appropriate confinement system. Mechanical/material systems are limited by the physical constraints mentioned above. Magnetic confinement systems are limited by the strength (density) of the magntice field - which is one of the critical contraints to fusion based on magnetic confinement.
  12. Jan 7, 2007 #11
    Hi Astronuc,

    Just to clarify, I wasn't talking about using a particle accelerator for action-reaction purposes. The accelerator would be to liberate energy from the Thorium by spallation (ie. an Accelerator Driven Thorium Reactor)

    The Thorium Reactor would then be used to heat a propellant like hydrogen, which would then be expelled at high speed.




    http://users.ox.ac.uk/~corp0935/nuclear.doc [Broken]

    A thorium-powered Earth-to-Orbit vehicle would have so much energy to spare that it could operate as a reusable single-stage unit, with airline-style operation and turnaround. Perhaps it could even fly to the Moon or even Mars in a single round trip.

    A more near-term goal could be to create a thorium-powered hypersonic bulk cargo transport, able to carry large amounts of cargo to any point in the world in a couple of hours, whether for military purposes or in the service of the global economy. As an unmanned or robot craft, it would be purely for the transport of cargo.
    Last edited by a moderator: May 2, 2017
  13. Jan 7, 2007 #12


    User Avatar
    Staff Emeritus
    Science Advisor

    I understand that - but if the thorium is solid, the heat has to be conducted from the thorium fuel to the hydrogen, and the system is thermally limited. Liquid fuel will creep and deform, and might lead to blocked coolant channel, which if continuing to melt results in disintegration of the core. Also, one still needs a power supply for the laser, which means some of the propellant must be bled off and passed into a power cycle.

    If the thorium is a vapor, intimately mixed with the hydrogen gas (propellant), that might work, but then one loses a portion (possibly a good portion) of the Th vapor with the fission products and hydrogen, which significantly increased the molecular mass of the propellant which reduces the specific impulse.

    I'll have to check the sources cited, but generally, I do not trust Newscientist.

    There have been gas core concepts, but all have been hypothetical, and based on my experience, I am skeptical about all gas core concepts.

    The Pluto core had to survive temperatures of ~2,500°F (1,600°C) and

    Seconds and minutes prove that the core configuration can go critical and hold power - but a viable system must operate for hours or days. And I'm sure this was highly enriched U, since it was a compact core.

    Accelerator driven systems (spallation based) are notoriously inefficient.

    Actually there was consideration of nuclear ramjets for propulsion in the atmospheres of Jupiter and Saturn. The craft would descend into the atmospheres, collect the gases and blast back into orbit. Well - maybe some day in the future.
    Last edited: Jan 7, 2007
  14. Jan 7, 2007 #13
    Well, if you had a pebble-bed design, this would provide a lot of surface area for contact with the propellant, in addition to increased safety.

    Or then how about a 2-state-to-orbit system, with conventional lower stage, and the 2nd being nuclear-electric, like VASIMR, MPD, or DS4G.

    I was even thinking, what if you used a giant blimp as your lower stage, just to lift you upto the aeropause, where you could then resort to nuclear-electric propulsion to achieve orbit?
  15. Jan 8, 2007 #14
    Another thing I was thinking about, was if the hydrogen were to combust with the airstream after being heated by the reactor, then you'd have steam which you could also pump heat into using a helicon microwave antenna.
  16. Aug 22, 2008 #15

    I'm the great-nephew of Father Michael DeLise Lyons, S.J. (Jesuit), who supplied the beryllium for our nuclear weapons from 1944 to 1974, when he died of chronic beryllium lung disease (Berylliosis). Yes, the final element of the Manhattan Project was supplied by a Jesuit priest.

    Father James Bernard Macelwane, S.J. of St. Louis University (famed father of the Seismology Society of America) also had "Q-Level" clearance and assisted the Manhattan Project in selecting sights for weapons factories and sylos.

    My question is whether you believe cars could be powered by Thorium?

    Jeff McQueen
    Rochester Hills, Michigan
  17. Aug 22, 2008 #16


    User Avatar
    Staff Emeritus
    Science Advisor

    Hello, Jeff, welcome to PF, and thanks for the history lesson.

    With respect to the question of thorium powered cars, if one is asking about a fission-based system - the answer would be know. Too much shielding! Small fissile cores require very high enrichments, and such core would not be practical for personal transportation.

    If one if referring to RTG's (radioisotopic thermal generators), Th-232 would not be practical either due to its long half-life. RTG's are typically based on Pu-238.
  18. Nov 12, 2008 #17
    Heating hydrogen by contact isn't interesting, whatever the energy source is. The reason is that the temperature you achieve is limited by technology to under 3000K, and then hot hydrogen isn't that much better than H20 at the same temperature.

    Once you've added a tank for the hydrogen and a reactor to get heat, you get a rocket performance that isn't so fantastic - not fantastic enough to develop a new technology.

    Anyway, once you're in orbit at about 1AU from the Sun, a solar furnace gives the same temperature (or better) to your hydrogen without the typical drawbacks.

    In other words: a really better rocket engine needs another way of accelerating the ejected mass.
  19. Nov 16, 2011 #18
    Can you use Thorium4+ like Xenon in hall thruster? Its heavier and thus may give a higher ISP?
    It also seems to be more easily attainable than Xenon?

    Then when you store it on-board the space craft it can provided heat, and can the decay be used to power a RTG type power plant?

    It seems like you can some serious multitasking out th+4.

    What do you guys think?
  20. Nov 27, 2011 #19
    About Rubbia's thorium sub-critical reactor: huge disappointment to me.

    I believe to understand it needs plutonium to start. Not 235U, not 233U, but plutonium - and it doesn't breed any.

    Then, if you put figures on the energy produced and the amount of plutonium needed, you get that each GW of Rubbia's reactor would require 10GW of uranium reactors, just to produce the plutonium that starts a Rubbia reactor.

    So it wouldn't be a sustainable model. Only a way to reduce the huge stock of plutonium Mankind has. Nice goal intrinsically, but for that, we have Mox burnt in PWR, and we can burn plutonium-thorium in some PWR, without a costly new development.
  21. Nov 27, 2011 #20
    Non-chemical rocket propulsion:

    Either you find impulse in-situ to produce a force. Solar sails do work, they're being developed eventually, and will bring excellent performance near our Sun - but are less usable to satellize around Saturn, say. Solar wind catchers are completely impractical up to now and as far as I know.

    Or you bring with you a mass you eject, and then the faster the ejection to save mass, the more power you need - better speed than chemical means an energy more compact than chemical, provided you bring it with you.

    The sub-variant is to harvest energy from the surroundings, like from Sunlight. Photocells and ion thrusters would cumulate inefficient steps and preclude any significant thrust, but if you heat hydrogen with Sunlight, you improve the ejection speed over chemicals (because hydrogen is lighter, not because of temperature), can obtain some thrust usable once in orbit, and the technology looks within grasp right now, if having the right ideas. Nice to go to Mars, to Jupiter as well, maybe beyond.

    If you like extracting energy from the atoms you eject, you might consider an existing proposal where an alpha emitter coats a propulsive area. All decay energy serves to eject alphas or nuclei, without losses. But as the layer is very thin to leave alphas through, it takes a huge area.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook