Title: Thermalization of Gauge Theory and Gravitational Collapse
1Thermalization of Gauge Theory and Gravitational
Collapse
SL, E. Shuryak. arXiv0808.0910 hep-th
2Basic elements of AdS/CFT
- In large Nc, strong coupling ? limit, string
theory in AdS5xS5 background is dual to N4 SYM - pure AdS background
AdS-Blackhole
z0
z0
thermalization
horizon zzh
z?
z?
N4 SYM at T1/(?zh)
N4 SYM at T0
3Gravity Dual of Heavy Ion Collision
- E.Shuryak, S.Sin, I.Zahed hep-th/0511199
- RHIC collisions produce debris consisting of
strings and particles, which fall under AdS
gravity - SL, E.Shuryak hep-ph/0610168 studied the falling
of debris and proposed to model the debris by a
shell(ignoring the backreaction of the debris to
AdS background)
4hologram of the debris
Q
Qbar
SL, E.Shuryak arXiv0711.0736 hep-th
5Gravitational Collapse Model
- Israel spherical collapsing in Minkowski
background.
6- Gravitational Collapse in AdS (backreaction
included)
boundary z0
AdS-Blackhole
shell falling
horizon zzh
pure AdS
z?
7Gauge Theory Dual
- gravitational collapse in AdS is dual to the
evolution of N4 SYM toward equilibrium - Different from hydrodynamics (locally
equilibrated) non-equilibrium is due to spatial
gradient. - Our model no spatial gradient. The SYM is
approaching local equilibrium.
8Israel junction condition
- continuity of metric on the shell
- matching of extrinsic curvature
- where
- Shell
- gij induced metric on the shell
9Falling of shell
-z0
Initial acceleration
Intermediate near constant fall
Final near horizon freezing
-zh
10- Physical interpretation of p, z0 and zh
- The parameter p should be estimated from the
initial condition on the boundary (energy density
and particle number) - z01/Qs1/1GeV zh1/(?T)1/1.5GeV
- Qs saturation scale
- zh initial temperature of RHIC
- The initial temperature of RHIC is determined
from initial collision condition
11Quasi-equilibrium
- axial gauge where ?z, t, x
- graviton probe where mt, x
- one-point function of stress energy tensor
- the same as thermal case
- Two-point function deviates from thermal case
-
12graviton probe h_mn
infalling
outfalling
horizon zmzh
infalling
AdS-BH (thermal) limit
13Graviton passing the shell
- matching condition given by the variation of
Israel junction condition - hmn outside and inside are continuous on the
shell - hmn outside and inside should preserve the EOM of
the shell
14Quasi-static limit
- Although the shell keeps falling, it can be
considered as static for Fourier mode - ?gtgt dz/dt
- NOTE
- the frequency ? outside corresponding to
frequency ?/f(zm)(1/2) inside
t_out
t_in
15Asymptotic ratio
- Starinets and Kovtun hep-th/0506184
- scalar channel hxy
- shear channel htx, hxw
- sound channel htt, hxxhyy, htw, hww
- where umzm2/zh2
- as um?1, f(um) ?0. Infalling wave dominates the
- outfalling one.
16Retarded Correlator and Spectral Density
- boundary behavior of hmn ? retarded correlator
Gmn,kl ? spectral density ?mn,kl
17spectral density ?mn,kldeviation from thermal
Rxy,xy
?
scalar channel q1.5 black um0.1, red um0.3,
blue um0.5, green um0.7, brown um0.9
18Rtx,tx
?
shear channel q1.5 black um0.1, red um0.3,
blue um0.5, green um0.7, brown um0.9
19Rtt,tt
?
sound channel q1.5 black um0.1, red um0.3,
blue um0.5, green um0.7, brown um0.9
20- spectral density
- the oscillation damps in amplitude and grows in
frequency (reciprocal of ?) as um ? 1. Eventually
the shell spectral density relaxes to thermal
one.
21- The WKB solution shows the oscillation of the
shell spectral density rises from the phase
difference between the infalling and outfalling
waves. - Further more, the frequency of oscillation in
spectral density (reciprocal of ?) corresponds to
the time for the wave to travel in the WKB
potential (Echo Time) - Echo Time approaches infinity as um ? 1
22Conclusion
- The evolution of SYM to equilibrium is studied by
a gravitational collapse model - Prescription of matching condition on the shell
is given by variation of Israel junction
condition. AdS-BH (thermal) limit is correctly
recovered - Spectral density at different stages of
equilibration is obtained and compared with
thermal spectral density. The deviation is
general oscillations. The oscillation is
explained by echo effect damps in amplitude and
grows in frequency, eventually relaxes to thermal
case.
23WKB solution(?gtgt1)
with
a_0, b_0, c_0 are functions of q, ?, u_m r
outfalling/infalling
24backup