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

Nuclear Power and Reactor Scale.

  1. Feb 21, 2005 #1


    User Avatar
    Gold Member

    Most commercial nuclear fission reactors are large, on the 1000MW order of magnitude.

    Some of the early experimental reactors were small, and the reactors in nuclear submarines are relatively small. I understand that the Navy even has one very small (crew of 7 people, 400 tons v. 7,000-9,000 for nuclear attack submarines) nuclear submarine that also moves very slowly 4 knots, which does searchs of the ocean floor (e.g. for wreckage). See here: http://www.chinfo.navy.mil/navpalib/factfile/ships/ship-nr1.html A couple of commercial ships of a freighter size were once nuclear, but have been converted back to conventional fuel.

    I've also heard about self-contained small nuclear reactors that they're talking about trying in an Alaskan village (on a pebble bed concept, IIRC). See http://www.adn.com/front/story/4214182p-4226215c.html

    What are the practical limits that impact the scale of a nuclear fission reactor? For example, what prevents someone from developing a nuclear powered airplane?
    Last edited: Feb 21, 2005
  2. jcsd
  3. Feb 21, 2005 #2
    Know Nukes on small and airborne reactors

    The Toshiba 4s.

    The Toshiba 4S is a sodium-cooled fast-neutron reactor that has virtually no moving parts and that for years can opertate without refueling. The pebble bed is a gas-cooled slow-neutron reactor with lots of moving parts and that needs to be refueled constantly and continuously.

    One frequently-cited factor is that heat engines tend to be more efficient when they are larger.

    Graham Cowen weighed in on this in the recent Better liquid oxide coolant thread on the Know Nukes list:

    > >No they wouldn't. Multiple separate rad shields
    > >on separate reactors would be needed, and each
    > >would have to be for real: decimetres of dense hydrogen
    > >plus decimetres of dense heavy metal, all blocking every
    > >ray path. Eight-pi shielding, or 12-pi, counting
    > >the multiplicity of reactors. A nuclear aircraft
    > >would be a thousand-tonne aircraft, maybe 2,000.
    > I misread the original post. You expect to fly a reactor!!!

    Zero or two or four, I think; not just one.
    One did fly ~50 years ago, once, as a passenger.
    The shield weighed 60,000 pounds IIRC;
    that's not for real.

    I think nuclear aircraft may fly if some worthwhile job
    is found that only aircraft of 2,000 tonnes or more can do,
    because airport and flight path neighbours will be more
    comfortable with them than with conventionally powered
    aircraft that big, i.e. with ~1,000 tonnes of kerosene or lH2,
    50 to 100 terajoules of potential BLEVE, on board.
    Wouldn't you be?

    In short, nuclear aircraft would certainly be safer than chemical aircraft, but they might have to be very large.
  4. Feb 22, 2005 #3


    User Avatar

    Staff: Mentor

    Right now the largest practical light water reactor (LWR) in commercial operation is the French N4 series. Civaux-1 is one of four of the new N4 nuclear reactor series. The other three being: Chooz B-1, Chooz B-2 and Civaux-2, each with a capacity of 1516 MWe (gross), 1450 MWe (net), with a thermal capacity of approximately 4270MWth. The core consists of 205 assemblies with an active fuel length of 4.27 m (14 ft).

    Some background - http://www.memagazine.org/backissues/aug98/features/reactor/reactor.html



    There is also the EPR (European Pressurized Reactor) Project. The EPR is a four-loop reactor designed for a thermal output of 4250 MWth and an electrical output capacity of 1500 MWe. Its core comprises 241 fuel assemblies, each containing 264 fuel rods and 81 control rods.

    Apparently the Russians are planning a VVER-1500 (1500 MWe) design as well.

    Nuclear powered aircraft would seem to be impractical from a thrust to weight ratio. The major impediment is the shielding.
  5. Feb 22, 2005 #4


    User Avatar
    Science Advisor


    The reactor core can be quite small, physically - but still put out a lot of
    power. The limit to how much power you get out of a reactor has more
    to do with how effectively you can cool it, rather than any limit as to
    how much the core can generate.

    The core of the research reactor at M.I.T. is about 15 inches in diameter
    and produces 5 Mwt:



    The core of the now shutdown University of Michigan reactor was a
    2-foot cube from which they extracted 2 Mwt. However, that core was
    essentially the core of a submarine reactor - and the 2 Mwt limit was
    only because that's how much heat the cooling system for the University
    of Michigan's reactor could extract. Put a cooling system with more
    capacity on it - and you can power a submarine.


    Although you could devise a reactor that could output the power of
    a few jet engines - the shielding necessary would make the plane too
    heavy. So a reactor-powered plane isn't practical.

    Dr. Gregory Greenman
    Last edited: Feb 22, 2005
  6. Feb 23, 2005 #5
    I'm not sure about this, but I've heard some pacemakers where powered by small amounts of plutonium
  7. Feb 23, 2005 #6


    User Avatar

    Staff: Mentor

    Plutonium-238 (non-fissile isotope) has been used to power scientific equipment in spacecraft and implanted heart pacemakers.

    See also Nuclear Batteries
  8. Feb 23, 2005 #7


    User Avatar
    Science Advisor


    Those aren't nuclear reactors. They are called RTGs for
    Radioisotope Thermal Generators.

    Whereas nuclear reactors derive their energy from the fissioning of
    fissile materials like U-235 and Pu-239; RTGs get their energy from the
    heat produced by radioactive decay of some radioisotope, most notably
    Pu-238, as Astronuc states.

    In an RTG, you let the radioisotope generate heat due to its radioactive
    decay, and you then convert the heat to electricity - a thermocouple is
    one way of doing that.

    The requirements for the "fuels" of reactors and RTG are different. The
    reactor requires a fuel like U-235 and/or Pu-239 which is fissile. The
    fuel doesn't have to be radioactive - in fact U-235 has a pretty low level
    of radioactivity.

    For the RTG, being fissile is immaterial - you want something that is
    quite radioactive so it produces a lot of heat. That's Pu-238.

    So nuclear reactors and RTGs are two very, very different animals. The
    RTGs are what power pacemakers.

    Dr. Gregory Greenman
  9. Mar 4, 2005 #8
    Nasa did some experimentation with using a reactor as a rocket engine... to my understanding the tests were successful as to whether it would work or not. I don't think it was practical for actual use because of the radiation it spewed out in the process... and what we send up tends to come back down again, sometimes unexpectantly. It could probably be used in deep space for power generation for an ion rocket engine... but it would need to be assembled there and kept a long way from
    the rest of the science package or crew.
  10. Mar 5, 2005 #9
    When did they test this? And how would you use a reactor as a rocket engine in the first place?
  11. Mar 5, 2005 #10
    Nuclear rocketry


    Nuclear rockets have, in one way or another, been studied for the past fifty years.

    You use the reactor-generated heat directly, or you use electricity generated by the reactor, or you use both of those things to launch propellant out of a rocket nozzle. For example, if the propellant is ionized hydrogen, you might use electrical energy to power electromagnets arrayed so as to accelerate the hydrogen ions. Some nuclear-rocket designs that use this mode of operation also use heat from the reactor to heat up the hydrogen propellant, and hence these designs use both heat and electromagnetics to provide energy to the propellant.

    The above link has more detail on nuclear rockets. Here is another one:

    There are may nuclear propulsion designs. One I am interested in is the vapor core (also known as gas core) reactor with MHD generator. Vapor (or gas) core means the fission fuel is normally in a vapor state instead of a solid or liquid state. This technology in a nuclear rocket might be very efficient since it allows the reactor to operate at a high temperature and because heat engine efficienvy tends to rise with rising temperature (per Carnot's law). Here is a link to a general description of a vapor core reactor with MHD generator:
    Last edited: Mar 5, 2005
  12. Mar 7, 2005 #11


    User Avatar
    Science Advisor


    Do a Google search on "NERVA" - Nuclear Engine for Rocket Vehicle Applications.

    Here's some of what I found:


    [This one has some good drawings of the NERVA engine when you click
    on the "Image Information" section. ]



    Dr. Gregory Greenman
  13. Mar 8, 2005 #12
    There is also MITEE, which is a rather recent design study on nuclear thermal propulsion. It can be found at this site:


    The weight of there main reactor design using U-235 is about 70 kg. What's more surprising is the whole engine weighs just 140 kg, puts out 14000 N of thrust, and has a specific impulse of 1000 seconds (over twice that of liquid oxygen/liquid hydrogen). They also have some other reactor designs using more exotic nuclear materials like U-233 and Am-242m, which shrink the reactor down to 40 kg and 25 kg, respectively.

    I too have wondered if a reactor could some how be miniaturized, small enough to power something like a car, boat, or small airplane. Maybe using a closed cycle with a small gas turbine for power generation? I had the idea looking over this site:


    It's a miniature turbo shaft engine. You would need a lot of radiation shielding though, wouldn't you? :grumpy:

    The real question is, how far can you miniaturize it? Like, could you make one the size of a desktop computer power supply, or a laptop battery? :tongue2:
  14. Mar 8, 2005 #13


    User Avatar

    Staff: Mentor

    Shielding is certainly an issue with any fissile system.

    Miniturization will be limited by critical mass at whatever enrichment one chooses. The pits in nuclear weapons are about as small as it gets. Am242 would allow a slightly small CM than Pu239. The key issues in a mini-reactor for vehicle transportation are:

    Control of the mini-reactor.
    Heat transfer.
    Disposition of fission products.

    No. The diameter of CMs of pure fissile materials are on the order of cm's, and then one must add shielding.
  15. Mar 8, 2005 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What if it crashes?

    What if it's hijacked?

    Nuclear reactors and subs are able to be reasonably isolated and kept secure from the general populace. Aircraft are not.
  16. Mar 8, 2005 #15


    User Avatar
    Science Advisor


    Nuclear propulsion would not be used in aircraft - nuclear propulsion is
    for SPACECRAFT!!!

    I believe that falls in the "reasonably isolated" category - even more so
    than subs.

    As far as crashing - NOBODY intends a nuclear propulsion system to be
    used at takeoff. The nuclear powered spacecraft would be lifted to
    Earth orbit via regular chemical rockets.

    Since the reactor won't have been started yet - the fuel will be no more
    radioactive than it was when first dug out of the ground. Therefore,
    a crash due to failure of the chemical rockets during boost phase would
    have minimal consequences.

    After the rocket is in orbit - and the nuclear engines are fired up - the
    rocket is not coming back down - it already has orbital velocity - and
    unless one were to use the nuclear engine as a retro-rocket - and nobody
    would ever do that - the craft is not coming back to Earth.

    Dr. Gregory Greenman
    Last edited: Mar 8, 2005
  17. Mar 8, 2005 #16
    I was wondering about that. Thanks for asking and clearing that up enigma and Morbius.
  18. Mar 8, 2005 #17


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Well, his original question was:

    and I was responding to that question with a few real-life reasons why it would be a patently bad idea.
  19. Mar 8, 2005 #18


    User Avatar
    Gold Member

    So does this follow because there are economies of scale in shielding? Or is it simply that anything short of a boat or submarine can't allocate a large enough proportion of vehicle weight to propulsion and shielding to make it work?
  20. Mar 8, 2005 #19


    User Avatar
    Science Advisor


    That's pretty much it. You have to have a vehicle big enough so the penalty
    that you pay in shielding is a small fraction of the vehicle.

    If you want to power an airplane - weight is a premium - and a reactor
    doesn't make much sense.

    However, if you are going to power an aircraft carrier - then the shielding
    is a very, very small part of the vehicle - and reactors are good powerplants
    for carriers.

    Even though Nimitz-class carriers are among the largest warships afloat ;
    they can out run a lot of the smaller non-nuclear powered vessels.

    Dr. Gregory Greenman
  21. Apr 28, 2005 #20
    Safety issues aside, could you take a steam railway locomotive, put a reactor core in the tender and feed coolant gases into the firebox? Is this still too small? Or is the idea just daft?
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?