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

Maxwell's Demon

  1. Sep 6, 2003 #1


    User Avatar

    I'm not sure if I got the name right, but anyways...

    Suppose we have a nanoscopically small, but very clever device. (and with modern technology, this sort of thing is becoming almost practical...) It is basically just two compartments, with a hole in between. Contained in the compartments, we have an energetic gas. The hole is controlled by a very clever valve mechanism, that lets particles through in one way only. We let the device run, and hey presto, by simple random motion, all of the gas is gathered in one of the compartments. At the push of a button, the hole can be used to drive a turbine from the diffusion of the gas, thus generating energy - and decreasing entropy.

    Obviously, such a device would break the laws of thermodynamics. So, why exactly would it not work?
  2. jcsd
  3. Sep 6, 2003 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Because this 'clever valve' can't exist.

    - Warren
  4. Sep 6, 2003 #3


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    The demon has to find out which corpuscles are moving fast and which are not. In gaining this information he increases his entropy, and voila! calculations show that the entropy the demon gains is equal to the entropy the gas system loses by his ministrations! No contradiction. This is a well-known result in modern thermodynamics.
  5. Sep 6, 2003 #4
    I have not been doing anything on Thermo for much time, but I will try anyway!...

    If the system is isolated, the temperature of the gas will increase, and the pressure too... If the temperature increases, the entropy will do too... Let's imagine that the gas is an ideal one (it couldn't be *good* since the system is nanoscopical, but I think we can do it to see what happen). We will have
    S2-S1=cLn(T2/T1)>0 because T2>T1, so no decreasing entropy...

    I bet for it...
    Last edited: Sep 6, 2003
  6. Sep 7, 2003 #5


    User Avatar
    Science Advisor

    Fascinating !
    So all this talk about thermodynamics being violated
    on tiny scales in devices like the one FZ+ talks about
    here is wrong/misinterpreted by the layman in terms of
    being able to use that energy ?

    Live long and prosper.
  7. Sep 7, 2003 #6
    Here are Maxwell's own original
    words in setting out this train
    of thought:

    James Clerk Maxwell
    Address:http://webserver.lemoyne.edu/faculty/giunta/demon.html [Broken]
    Last edited by a moderator: May 1, 2017
  8. Sep 9, 2003 #7


    User Avatar
    Science Advisor

    Interestingly, Maxwell made no speculations re the entropy gained by the demon, but only posited the argument that the 2nd law applies to large numbers of particles.

    A much simpler system was proposed by Feynman, where the demon is essentially just a ratchet. A 'windmill' attached to the ratchet would allow for thermal fluctuations to apply forces of either direction to the windmill, but the ratchet would only allow motion in one direction, thus seeming to violate the 2nd law. The resolution is the fact that for the ratchet to be this sensitive, the pawl of the ratchet must be microscopic as well, and it would itself be subject to thermal fluctuations; essentially, it would bounce around, and the ratchet wouldn't work.

    Recently, the above was put on a firm basis. See Phys.Rev.E 59, 6448 (1999). Link is

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook