6
$\begingroup$

The SHO in QM with mass $m=1$ has action $$ S[x] = \int dt \frac{1}{2} \dot x^2 + \frac{1}{2}\omega^2 x^2 $$ by integration by parts we see this is the same as 1 dim Klein Gordon QFT action with field $x(t)$ and mass $m=\omega$: $$ S[x] = \int dt \frac{1}{2} x ( -\partial_t^2 + \omega^2 )x $$ Now, as is done in section 2 of this article (http://authors.library.caltech.edu/8383/1/BOOejp07b.pdf) , we can then derive the 1+0D QFT Feynman propagator in the QFT: $$ \langle 0|\mathcal{T}x(t_i) x(t_f) | 0 \rangle = \frac{1}{2\omega} e^{-i\omega|t_i-t_f|} $$ Since the actions are the same, I feel like I should somehow be able to relate this result to the QM amplitude $$ \langle x_f,t_f|x_i,t_i\rangle $$ which is quite a complicated formula with sines and cosines as derived from the path integral in this article (http://web.mit.edu/dvp/www/Work/8.06/dvp-8.06-paper.pdf). Is there a neat way to map this 1D QFT to the QM SHO? Or from the QM SHO to the 1+0D QFT?

$\endgroup$
5
  • 2
    $\begingroup$ The actions are not the same! In the QFT case you have an integral over position too. In general QM is a "0D QFT" and doesn't have nearly enough information to describe actual QFT. $\endgroup$
    – knzhou
    Commented Apr 20, 2016 at 17:00
  • $\begingroup$ Well not if it is a 1D QFT: if one defines a 1 dim QFT, it should only depend on one parameter ( in my case i call this $ t $ ) , and then I define one bosonic field which I call $ x $, dependent on $ t $, i.e. $ x(t) $ is my bosonic field, living in 1 dimension (t). The action then integrates over the base space of the quantum fields, and that is just $ t $. So I'm pretty sure this is the correct 1D QFT action for a Bosonic field. $\endgroup$ Commented Apr 20, 2016 at 20:18
  • $\begingroup$ Sorry, I should have been more clear. The QHO is a "0+1 dimensional" QFT, with one time dimension. You want to match it up to a "1+1 dimensional" QFT, which doesn't work. $\endgroup$
    – knzhou
    Commented Apr 20, 2016 at 20:29
  • $\begingroup$ It's easier to see the difference in the classical limit. A classical harmonic oscillator is defined by a function $x(t)$, there is only one parameter at each time. A classical 1D field is defined by a function $\phi(x, t)$, which is totally different. QFT is much more complex. $\endgroup$
    – knzhou
    Commented Apr 20, 2016 at 20:30
  • $\begingroup$ Okay I'm a bit confused. A classical 1D field should just have a base space of just 1 coordinate right? Not 2. In the same way that 4D fields depend on 4 coordinates. The $ x (t) $ is my bosonic field, so I redefining $ x \to \phi $, I just have 1D field $ \phi(x) $ obeying the action above. EDIT: when I say dimension, I mean $ either $ spatial or temporal $\endgroup$ Commented Apr 20, 2016 at 20:53

1 Answer 1

3
$\begingroup$

Yes, it is possible to match the QM with a QFT in 1+0 dimensions. However, the Fock vacuum $|0\rangle$ (which is annihilated an annihilation operator $a|0\rangle=0$) is naturally related to the coherent states $$\hat{a} |z\rangle~=~ z|z\rangle \tag{1}$$

rather than position eigenstates $$\hat{q} |q\rangle~=~q|q\rangle.\tag{2}$$ [Of course, it is possible to translate between the different operators and eigenstates (1) $\leftrightarrow$ (2).] Therefore it is most easy to consider the coherent state overlap $$\langle z_f^{\ast},t_f | z_i, t_i \rangle \tag{3}$$ rather than the position space overlap $$\langle q_f,t_f | q_i, t_i \rangle \tag{4} .$$ A QFT 2-pt function $$\langle 0^{\ast} | T[\hat{a}(t_1) \hat{a}^{\dagger}(t_2) ]|0 \rangle \tag{5}$$ is closely related to the coherent state overlap (3) with two extra insertions and $z_i=0=z_f$. See e.g. Ref. 1 for details.

References:

  1. L.S. Brown, QFT; Sections 1.7-1.8.
$\endgroup$
1

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.