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

Propagator using Functional QFT

  1. Dec 29, 2012 #1

    I am trying to write down the propagator for a scalar field theory, but I want to try and get it in the functional representation. My plan is to compute the following:
    \langle \psi (x', t') | \psi (x,t) \rangle
    which gives the amplitude to go from x' to x. Now I guess I have to interpret this state as the ground state of the scalar field, since next I want to drop in a complete set of states
    \langle \psi (x', t') | \psi (x,t) \rangle = \int \mathcal{D}\phi \, \langle \psi (x', t') |\phi \rangle \langle \phi | \psi (x,t) \rangle = \int \mathcal{D}\phi \, \psi^{' *}[\phi] \, \psi [\phi]
    Is this procedure correct so far? Can I assume the wave functional is the ground state of the field theory in order to continue? Thanks.
  2. jcsd
  3. Dec 29, 2012 #2
    Wait, I think I got it. For a free scalar field the propogator would be like
    \langle 0 | \varphi (x) \varphi(x') | 0 \rangle
    Then we put in a complete set of eigenstates
    \int \mathcal{D}\phi \, \langle 0 | \varphi (x) | \phi \rangle \langle \phi | \varphi(x') | 0 \rangle = \int \mathcal{D}\phi \, \phi(x) \, \phi(x') \, \psi^{*}_{0}[\phi] \, \psi_{0}[\phi]
    Next would be to explicitly calculate what [itex]\psi_{0}[\phi] [/itex] would be and then do the functional integral I believe.
  4. Dec 31, 2012 #3
  5. Dec 31, 2012 #4
    Ah, the generating functional approach, I am trying to stay away from that right now. I want to compute the two point correlation function for a free scalar field theory using "wave functional" representation. I'm still trying to make it work though . . .

    currently, I am working with the lowering operator
    a(\vec{k}) = \int d^3 x \, e^{-i \vec{k}\cdot \vec{x}}(\omega(\vec{k})\varphi(x) + i\pi(x) )
    and solving for [itex]\Psi_{0}[\tilde{\phi}][/itex] using
    a(\vec{k}')\Psi_{0}[\tilde{\phi}]=\omega(\vec{k})\tilde{\phi}(\vec{k}')\Psi_{0}[\tilde{\phi}]+\frac{\delta \Psi_{0}[\tilde{\phi}]}{\delta \tilde{\phi}(\vec{k}')}=0
    and I get
    \Psi_{0}[\tilde{\phi}] = N \exp \left[-\frac{1}{2} \int d^3k \, \tilde{\phi}(\vec{k})\omega(\vec{k}) \tilde{\phi} ( \vec{k} ) \right]
    Now I am in the process of solving the gaussian integral
    \int\mathcal{D}\tilde{\phi} \, \tilde{\phi}(\vec{k})\tilde{\phi}(\vec{k}') \, \Psi_{0}^{*}[\tilde{\phi}] \, \Psi_{0}[\tilde{\phi}]
    but I can't seem to get it to work yet . . .
    Last edited: Dec 31, 2012
  6. Jan 1, 2013 #5
    it does not seem that it will work.Are you sure with it?
  7. Jan 2, 2013 #6
    You have to be careful here. You're probably using a coherent-state basis, which is a basis corresponding to eigenvalues of the field operators. This basis is overcomplete, so you need a compensating factor for this. Look at for instance Altland and Simons.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Propagator using Functional QFT