59. Eye on Arxiv: Quantum corrections to eta/s

I decided to turn the de Sitter discussion off for a moment, it would be more appropriate to return to it after the paper is ready anyway Hopefully, it will not take too much time for you to see it in ArXives.

Today I want instead to briefly discuss the paper “Quantum corrections to eta/s” by Myers et al.

As many of you are well aware of, in principle, AdS/CFT correspondence allows to understand (in the very first, rough, approximation) the non-equilibrium properties of the quark-gluon plasma. When I mention the word “non-equilibrium”, I actually do not mean dynamical properties of the plasma far away from the thermal equilibrium, but the regime where linear response analysis can be actually applied.

Although applying linear response to the strongly coupled QG plasma naively does not sound very physical, the nature turns out to work in such a way that very soon after the collision of two heavy nuclei (Au-Au, for example) dynamical equilibrium is reached in the sense that effective equation of state in the QG plasma is well settled down towards its equilibrium value. Although the real thermal equilibrium is very far from being achieved, the QG plasma is in the state which can be called “prethermalization”. In practice, prethermalization means that only UV part of the overall spectrum is relatively far from the thermal one, but we do not care about it much since IR low energy part interesting for us is well under control (for example, spectrum is power law like in turbulence or exsibits thermal behavior with some effective temperature , which is a slowly varying function of scale).

Anyway… in the “prethermalization” state one can apply linear response to understand behavior of non-equilibrium QFT correlators. This behavior is very far from being trivial because a) we discuss the state with very large occupation numbers and b) theory is effectively at strong coupling (we are talking about length scales ).

The latter issue however can be treated by means of AdS/CFT; namely, we map strongly coupled QG plasma to gravity degrees of freedom propagating on the AdS background. The fact that we are dealing with conformal SYM instead of confining YM without any SUSY and number of colors equal 3 somehow does not bother us much because:

1) plasma is at large temperature, where scaling behavior effectively looks similar to CFT one
2) well, corrections are actually important but let us forget about them for a moment and see how far we will be able to advance

Applying AdS/CFT immediately gives rise to a very interesting result: so called shear viscosity bound. Namely, as it turns out, for any gauge theories with an Einstein gravity dual in the limit of a
large number of colours and large t Hooft coupling

(1)

and this result is claimed to be universal and even the lowest possible value of the shear viscosity in Nature (liquid He for example has larger shear viscosity).

As I said, finite ‘t Hooft and corrections are important, but their calculation is very far from trivial. In 2004 Starinets, Buchel and Liu have found the leading correction -

(as you see, it is of the weak power law form, so is hardly negligible in practice; nevertheless, experiments show that real QCD QG plasma comes very close to the bound (1), so it is very interesting to understand why this happens).

In Myers et al. paper the leading correction is calculated for the first time. One has:

The result is calculated of course by writing higher order corrections to the Einstein-Hilbert action (as you may remember, corrections correspond to corrections on the string theory side, while – to higher order terms in the Einstein-Hilbert action).