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

Nuclear fusion (requirements to start fusion)

  1. Nov 11, 2011 #1
    I have a question , hoq much pressure s needed for atoms to start to fuse(hydrogen atoms i guess) lets say we use the same tritium and deuterium and we put them in an environment with pressure how much pressure would it need? could you please say the pressure like in G like in gravitational measurment?
    Also I would like to know if pressure alone can start the fusion reaction or does it need pressure and heat but if so then how would that reaftion could ever start because heat comes after the reaction has started and it would sound logically that the pressure comes first the atoms fuse and create heat in the process?

    I guess its the "who was first, the egg or chicken" question

    let's imagine a centrifuge that spins so fast that at the outr side of the centrifuge such pressure is built up that is sufficient to start fusion, the fusion starts but the centrifuge is surrounded by water that absorbs the heat and turns to steam? would the fusion reaction continue if the centrifuge would continua to spin at the same rate and the pressure would be the same only the heat would never build up like in a typical fusion reaction that would destroy the apparatus?
    So the basic questions is is pressure enough or do we need pressure and heat combined to fuse atoms?

    Thanks for answering:)
  2. jcsd
  3. Nov 11, 2011 #2


    User Avatar

    Staff: Mentor

    Welcome to the PF.

    The Lawson criterion is what you are asking about. It uses plasma density, confinement time and temperature as a figure of merit for ignition:


  4. Nov 11, 2011 #3
    A glass of water has deuterium and tritium nucleii fusing (a few of them, but they are there), so the question actually has no answer unless you talk about getting enough nucleii to fuse often enough to beat some other rate - which leads you to different criteria like Lawson's.

    Heating rates are basically proportional to density squared times a reaction rate that scales very, very steeply with temperature and is negligibly small unless you get into the tens to hundreds of million degree range. That's before you start self-heating, so you have to get there some other way.
  5. Nov 12, 2011 #4
    Yes first time , thanks for welcoming:)
    So for a fusion reaction to be self sustaining it has to have either
    1) huge enough temperature and pressure or just temperature? i guess both are needed right?

    But a fusion reaction is achievable under different condition right, like under huge pressure?

    Okay so my question is if a huge pressure like huge centrifugal pressure from a very high rpm turbine is applied then would the fusion reaction keep on going even without the extreme heat ?
    because imagine that the heat would go away in form of steam from water cooling but to keep the reaction going the pressure would be kept very high so that new and new atoms start to fuse, could someone explain is this possible?and if is would the energy put in to keep the pressure be bigger than the energy extracted from the heat of the reaction or not?

    I guess i could answer my question myself with numbers and formulas but that aint so easy so could you please give a lil light on this? :)
  6. Nov 12, 2011 #5
    A self-sustaining fusion reaction needs density and temperature in some combination - you could write this as pressure and temperature, or pressure and density, but that seems generally pointless as it's conceptually simpler to think of density and temperature, which is what the Wiki link uses (did you read that?).

    High pressure generally = high temperature, and high density and off-scale high temperature generally = ridiculously high pressure, so good luck with that centrifuge idea. Self-sustaining fusion schemes are in the tens-of-millions of degree range, as I pointed out and as the Wiki link also shows (did you read that?).
  7. Nov 13, 2011 #6
    not that i would think of it as a idea that has a chance for real life more like theoretical interest in things, yes i read the article in wiki.
  8. Nov 13, 2011 #7
    Why do you want a fusion reaction?
  9. Nov 13, 2011 #8
    It's not so much about what I or we want or can it's more about what is possible, there are some things that you have or can find out practically and some that can be theorized, this is one of them.
    I guess the problem about fusion reactions is that they produce too much heat, well ofcourse we want the heat for steam that could be used in turbines, but again too much heat is no good, if not for the heat I think and some calculations show that lets say a 10 m in diametter centrifuge with some 30 000+ rpm could get H atoms dense enough at the outer walls to make them fuse, because centrifugal force acts just the way gravity acts so if we know that gravity and the huge gravitational pressure causes atoms to fuse in center of sun and not only , this should do the trick but again there is this huge heat everywhere wich makes everything so bad...
  10. Nov 13, 2011 #9
    I think you have a lot of very basic misconceptions about all this, perhaps best to start with a basic book about power production and physics - I don't think there is much that a forum thread can do for you.
  11. Nov 13, 2011 #10


    User Avatar
    Staff Emeritus
    Science Advisor

    Crazymechanic, you need to isolate the plasma from the walls of the containment vessel. The heat requirements are enormous and requires something like magnetic fields to keep the plasma from hitting the walls of the reactor.

    Also, there is a difference between pressure and heat. The heat inside the sun occurs because the plasma has collapsed under a great amount of pressure. This generates the heat needed to fuse and also keeps the plasma dense enough for the fusion reactions to continue at a reasonable rate. (It does no good if you have the temperature requirements for fusion but your plasma is only dense enough so that 10 reactions per second occur)

    I highly doubt the kind of pressure needed for confinement is capable of being mechanically generated by spinning the reactor around.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook