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

Green Functions and BC's

  1. Jun 26, 2008 #1
    Hey folks,

    I'm trying to get a handle on my old Nemesis, Green functions. I have a massless scalar field confined between two parallel plates separated by a distance a (in the z dimension) and the field satisfies Dirichel BC's. Basically I'm trying to work from line 1 of a book to line 2 (K. Miltons the Casimir Effect p23).

    'The Green function satisfies'


    "We introduce a reduced Green function g(z,z) according to the Fourier Transform"

    [tex]G(x,x')=\int\frac{d^dk}{(2\pi)^d}e^{i\vec{k}.(x-x')}\int\frac{d\omega}{2\pi}e^{-i\omega(t-t')}g(z,z') [/tex]

    This is all the book says so sorry of thats not much info. I'm fairly sure that [tex]\partial=\nabla+\frac{d}{dt}[/tex].

    What I want to understand (and see the math for) is how to get from line 1 to line 2. I'm pretty sure that it involves fourier transforms, but I would like to see it. Also, I don't understand the concept of a reduced green function. Can anyone either point me to a good reference, or better still explain how and why it is used.

    I hope someone can walk me through this.

  2. jcsd
  3. Jun 27, 2008 #2

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    I don't know all the details of that particular problem, but what I remember from electrostatics is that the "reduced" Green's function is essentially a separation-of-variables technique. If we have, for some linear operator L,

    [tex]LG(\vec x, \vec x') = \delta^3(\vec x - \vec x')[/tex]

    then we can write

    [tex]\delta^3(\vec x - \vec x') = \delta(x - x') \, \delta(y - y') \, \delta(z - z')[/tex]

    (or we could use some other coordinate system, such as [itex]r, \phi, \theta[/itex], transforming to it with the proper Jacobian).

    Given the boundary conditions, x and y are free, but z is bounded by the planes z=0 and z=a. So we can write

    [tex]\delta(x - x') = \frac{1}{2\pi} \int dk_x \, e^{ik_x(x-x')}[/tex]

    which is just an identity using Fourier transforms (the integral is over the entire real line). We can do likewise in y, because y also has no boundaries. This leaves us:

    [tex]LG = \frac{1}{(2\pi)^2} \int dk_x \, e^{ik_x(x-x')} \, \int dk_y \, e^{ik_y(x-x')} \, \delta(z - z')[/tex]

    My memory gets shaky at this point, but you might be able to see where to go. For [itex]L = \partial^2[/itex], L is pretty simple to invert for the two Fourier transforms. What's left is a function [itex]g(z, z')[/itex] that needs to be solved for.
  4. Jun 27, 2008 #3
    Thanks Ben. I think I actually realize what I don't understand now. Lets say our Geometry is in x,z and t. We can use techniques from Separation of variables to write our GF as:


    Where the [tex]\mu[/tex]superscript runs over x,z,t.

    Ok up to here, but then the second line of my first post implies that


    Ok, so here lies my problem. Why can I just assume that the reduced green function in x and t is just a delta function?
  5. Jun 27, 2008 #4

    Ben Niehoff

    User Avatar
    Science Advisor
    Gold Member

    I may have made a slight mistake; I'm not sure. But you should have

    [tex]\partial^2 G = \delta(x - x') \, \delta(y - y') \, \delta(z - z') \, \delta(t - t')[/tex]

    rather than just G on the left hand side. Once you have that, you can choose to represent some of those delta functions by their Fourier transforms (or Bessel function series, or what-have-you) in the frequency domain.

    The reduced Green's function in a particular dimension is not a delta function; it is [itex]\partial^2g[/itex] that is a delta function.
  6. Jun 27, 2008 #5
    ok, I think it should be:

    [tex]G(x,xs')=\int\frac{d^dk}{(2\pi)^d}e^{i\vec{k}.(x-x')}\int\frac{d\omega}{2\pi}e^{-i\omega(t-')}g(z,z') [/tex]

    because the physics is contained in the z direction so we just expect plane wave solutions in the x and t so we can replace them with delta functions. Then we put this expression for G into the first equation G''.

    Thanks for your insight! :)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook