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

Gibbs sum for particle

  1. Nov 24, 2006 #1
    a) Consider a system that may be unoccupied with energy zero or iccuped by one particle in either of two states, one of zer oeenrgy and one of energy epsilon. Show taht the Gibbs sum for this system is

    [tex] z = 1 + \lambda / \lambda\exp(-\epsilon/\tau) [/tex]

    b) Show that the thermal average occupancy of the ssytem is
    [tex] <N> = \frac{\lambda + \lambda\exp(-\epsilon/\tau)}{z} [/tex]

    c) show that the thermal average occupancy of the state at eneryg = epsilon is
    [tex] <N(\epsilon)> = \lambda\exp(-\epsilon/\tau)/z [/tex]

    d) Find an expression for the theram laverage eneryg of the system
    e) Allow the possibility tat the orbital at 0 and at epslon may be occupied each by one particle at the same time, show that
    [tex] z = 1 + \lambda + \lambda\exp(-\epsilon/\tau) + \lambda^2 \exp(-\epsilon/\tau) = (1+ \lambda) [1 + \lambda \exp(\epsilon/\tau)] [/tex]

    I will post my attempted solutions in a seaparate post.
    Last edited: Nov 24, 2006
  2. jcsd
  3. Nov 24, 2006 #2
    The Gibbs Sum or grand sum is given by

    for this problem at least...
    [tex] z = \sum_{N=0}^{1} \sum_{s(0)}^{s(1)}} \exp\left(\frac{N\mu-\epsilon_{s(N)}}{\tau}\right) [/tex]

    which comes out to

    [tex] z = \sum_{N=0}^{1} \exp(\frac{N\mu}{\tau}) + \exp\left(\frac{N\mu-\epsilon_{s(N)}}{\tau}\right) [/tex]

    whic his

    [tex] z = \exp(0) + \exp(\frac{\mu}{\tau}) + \exp(\frac{\mu-\epsilon}{\tau} [/tex]
    a term is missing because there is NO energy contribution by the s(0) state.

    also we defined lambda like [itex]\lambda = \exp(\mu/\tau}[/itex]

    [tex] z = 1 + \lambda + \lambda \exp(\epsilon/\tau) [/tex]

    for b) similar concept applies

    c) What is the diff between <N> and <N(epsilon)>?? Is N(epsilon)_ the number of particles whose energy actually contributes??

    d) [tex] <\epsilon_{s(N)}> = \frac{\sum_{N=0}^{1} \sum_{s(0)}^{s(1)} \epsilon_{s(N)} \exp((N\mu-\epsilon)/\tau)) }{z} [/tex]

    [tex] <\epsilon_{s(N)}> = \sum_{N=0}^{1} \frac{\epsilon_{s(N)}\exp(\frac{N\mu-\epsilon}{\tau})}{z} [/tex]

    [tex] <\epsilon_{s(N)}> = \exp(\frac{\mu-\epsilon}{\tau})}{z} = \lambda\exp(-\epsilon/\tau) [/tex]

    e) i Dont quite understand

    are they saying that we should accomodate N=1 only?? And N=0 is no more valid
    Last edited: Nov 24, 2006
  4. Nov 27, 2006 #3
    c) <N> is the average number of particles in the system, all energies included. <N(epsilon)> is the average number of particles with the particular energy epsilon in the system.
  5. Nov 27, 2006 #4
    d) The possible combinations are N=0, E=0; N=1, E=0; N=1, E=epsilon. You are on the right track in your first two attempts, so I think you should be able to figure it out from here.
    Last edited: Nov 27, 2006
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook