36. Eye on ArXiv: 6 May 2008 - Where does the cosmological perturbation theory actually break down?

Posted by Dmitry
May 6, 2008

I am going to briefly discuss today the paper by Cristian Armendariz-Picon, Michele Fontanini, Riccardo Penco, and Mark Trodden “Where does cosmological perturbation theory break down?”

The issue of possibility to describe inflation in terms of effective QFT keeps bothering people’s minds :) Let me remind you that two weeks ago we had a paper by Steven Weinberg discussing the similar subject. What is the essence of the problem that Christian et al. found? Naively, it seems that as long as H\ll{}M_P, dynamics of inflaton + metric perturbations can be described in terms of QFT (canonically normalized scalar field with some potential). Well, not quite so, as the authors say.

If you start to take effective QFT tree level corrections into account, the correlator for Mukhanov variable will have the form

\langle{}v(k) v(k')\rangle=\frac{1}{2k}\left(\left(1+\alpha_{20}\frac{H^2}{M_P^2}+\cdots\right)+\right.

\left.\left(\alpha_{22}\frac{H^2}{M_P^2}+\alpha_{42}\frac{H^4}{M_P^4}+\cdots\right)\frac{k^2}{M_P^2}+\cdots \right) (1)

etc., so the actual smallness parameter in the effective QFT expansion should not be k^2/M_P^2, but something like k^2\cdot{}H^2/M_P^4 instead.

I believe this fact by itself is not new. Take a correlation function for curvature perturbation \langle \zeta(k)\zeta(k')\rangle. Apart from the standard k^{-3} term there are corrections to this correlator small with respect to the parameter

H^2/M_P^2{}N (2)

where N is the number of efoldings (see our paper for example as well as recent works by David Seery, David Lyth etc.) These corrections grow while the number of efoldings grows and the expansion breaks down sooner or later. Now, if we take a look at the formula (1), we see that the smallness parameter there looks remarkably similar to (2). Indeed, the number of efoldings is defined as

N=-\frac{8\pi}{M_P} \int \frac{d\phi}{\sqrt{\epsilon}},

where \epsilon is the slow roll parameter.

What is the physical meaning of the effective QFT breakdown the authors found? The point is that the number of efoldings cannot be very large, and if it is, then inflation enters eternal self-reproducing regime. There the bare notion of homogeneous isotropic FRW backround breaks down (eternally inflating Universe consists of different Hubble patches, and the distribution of \phi among them is highly inhomogeneous). Still, the theory is under control even in this regime (instead of effective QFT, we can describe dynamics of the inflaton field by stochastic formalism).

By the way, authors say on the p. 7 that by in-vacuum they mean Bunch-Davies. This is probably a misprint: in-vacuum corresponds to the absence of particles at t\to\infty, while Bunch-Davies is Euclidean vacuum corresponding to the linear combination of in- and out-modes.

P.S. There is also very nice description of the Schwinger-Keldysh formalism in the paper.

Rate this:
2.5
These icons link to social bookmarking sites where readers can share and discover new web pages.
  • StumbleUpon
  • del.icio.us
  • bodytext
  • Reddit
  • Technorati
  • Facebook
Cosmology, Journal club, Master Postdoc, Quantum field theory

If you enjoyed this post, please consider to leave a comment, subscribe to the RSS feed or get new posts sent directly to your Inbox. If you are interested to know what I am doing right now, follow me on Twitter. The posts below are probably related to the subject of this one:

38. And again about effective QFT for inflation
53. Eternal inflation: stochastic approach 1 (Inflationary perturbations 7)
32. Eye on ArXiv: 30 Apr 2008 - Curvature perturbation from false vacuum inflation
48. Planck 2008: day 3
47. Planck 2008: First and second days

Comments

No comments yet.

Leave a comment


Science Blogs - Blog Catalog Blog Directory