Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Pressure of deconfined QCD for all temperatures and quark chemical potentials

A. Ipp, K. Kajantie, A. Rebhan, and A. Vuorinen
Phys. Rev. D 74, 045016 – Published 17 August 2006

Abstract

We present a new method for the evaluation of the perturbative expansion of the QCD pressure which is valid at all values of the temperature and quark chemical potentials in the deconfined phase and which we work out up to and including order g4 accuracy. Our calculation is manifestly four-dimensional and purely diagrammatic—and thus independent of any effective theory descriptions of high temperature or high density QCD. In various limits, we recover the known results of dimensional reduction and the hard dense/thermal loop (HDL/HTL) resummation schemes, as well as the equation of state of zero-temperature quark matter, thereby verifying their respective validity. To demonstrate the overlap of the various regimes, we furthermore show how the predictions of dimensional reduction and HDL resummed perturbation theory agree in the regime Tgμ. At parametrically smaller temperatures Tgμ, we find that the dimensional reduction result agrees well with those of the nonstatic resummations down to the remarkably low value T0.2mD, where mD is the Debye mass at T=0. Beyond this, we see that only the latter methods connect smoothly to the T=0 result of Freedman and McLerran, to which the leading small-T corrections are given by the so-called non-Fermi-liquid terms, first obtained through HDL resummations. Finally, we outline the extension of our method to the next order, where it would include terms for the low-temperature entropy and specific heats that are unknown at present.

  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
  • Figure
14 More
  • Received 20 May 2006

DOI:https://doi.org/10.1103/PhysRevD.74.045016

©2006 American Physical Society

Authors & Affiliations

A. Ipp

  • ECT*, Villa Tambosi, Strada delle Tabarelle 286, I-38050 Villazzano Trento, Italy

K. Kajantie

  • Department of Physics, University of Helsinki, P.O. Box 64, FI-00014 Finland

A. Rebhan

  • Institut für Theoretische Physik, Technische Universität Wien, Wiedner Hauptstr. 8-10, A-1040 Vienna, Austria

A. Vuorinen

  • Department of Physics, University of Washington, Seattle, Washington 98195, USA

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand
Issue

Vol. 74, Iss. 4 — 15 August 2006

Reuse & Permissions
Access Options
Author publication services for translation and copyediting assistance advertisement

Authorization Required


×

Images

  • Figure 1
    Figure 1
    The one-, two-, and three-loop two-gluon-irreducible graphs of QCD. The wavy line stands for a gluon, the dotted line a ghost, and the solid line a quark.Reuse & Permissions
  • Figure 2
    Figure 2
    Classes of IR sensitive vacuum graphs contributing to the QCD pressure at order g4. The black dots represent the one-loop gluon polarization tensor given in Fig. 3a and the indices ni stand for the numbers of loop insertions on the respective lines.Reuse & Permissions
  • Figure 3
    Figure 3
    (a) The one-loop gluon polarization tensor Πμν(P) divided into its vacuum (T=μ=0) and matter (vacuum-subtracted) parts. (b) The IR-safe Vac-Vac diagram contributing to the pressure at O(g4). (c) The IR-safe Vac-Mat diagram contributing to the pressure at O(g4). (d) The remaining “matter” ring sum.Reuse & Permissions
  • Figure 4
    Figure 4
    Comparison of the g4lng and g4 terms of the numerical computation and the analytic DR result, for various values of μ/T. The perturbative terms are subtracted up to order g3.Reuse & Permissions
  • Figure 5
    Figure 5
    Comparison of the HTL+ pressure and our numerical result pIV in the region of T=τgμ, τ=0.2, with the known perturbative terms from dimensional reduction subtracted and the entire quantities divided by g9/2. This plot shows that both the HTL+ result and our numerical one are accurate at least up to order g9/2. The renormalization scale has been varied between μ and 4μ. While pIVpDR is scale independent, pHTL+pDR has a scale dependence at order g4μ2T2g5μ4.Reuse & Permissions
  • Figure 6
    Figure 6
    Same as Fig. 5 but normalized to g5. While the HTL+ result is no longer accurate to this order and diverges logarithmically, our numerical result still correctly reproduces the dimensional reduction result for the pressure at order g5μ4.Reuse & Permissions
  • Figure 7
    Figure 7
    Thermal contribution to the interaction pressure δp as a function of T/mDT=0 for fixed chemical potential μ and coupling g=0.1. For this value of the coupling, the results of the numerical evaluation of panl+pringsafe and HTL+ coincide within plot resolution. The result is compared to the dimensional reduction pressure at orders g2, g3, g4, and g5 (where the latter is included only for completeness, as neither pIV nor pHTL+ contain contributions of order g5). The effect of varying the renormalization scale ΛMS¯=μ4μ is not visible for this value of the coupling.Reuse & Permissions
  • Figure 8
    Figure 8
    Same as Fig. 7 but for g=0.5. The results of the numerical evaluation of panl+pringsafe and HTL+ can now be distinguished due to their different content of higher-order terms. When two lines of the same type run close to each other, they differ by changing the renormalization scale ΛMS¯=μ4μ.Reuse & Permissions
  • Figure 9
    Figure 9
    Same as Fig. 8, but with pIV separated into panl and pringsafe. As the g4 contribution in δpanl only amounts to a small correction (of effective order g6), the shape of the full pressure curve as a function of T (beyond the rather trivial g2 contribution) is mainly determined by pringsafe. The renormalization scale dependence ΛMS¯=μ4μ is entirely due to panl.Reuse & Permissions
  • Figure 10
    Figure 10
    Same as Fig. 7, but for g=1. At this value of the coupling, the numerical result for pIV begins to be visibly affected by the choice of the magnetic mass (3.24) which here is taken with cf=1.Reuse & Permissions
  • Figure 11
    Figure 11
    Plot of the last term in Eq. (4.2) (dashed line) in comparison with δpHDL+ with the first two terms of Eq. (4.2) subtracted, in units of mD4=g¯4μ4/π4 and as a function of τ¯.Reuse & Permissions
  • Figure 12
    Figure 12
    Comparison of our numerical result pIV and the Freedman-McLerran result. The plot shows agreement to order g5, i.e., agreement in the nonvanishing coefficients up to and including order g4, and the absence of order g5 contributions.Reuse & Permissions
  • Figure 13
    Figure 13
    Same as Fig. 12, but divided by g6 instead of g5, revealing that pIV contains a term of order g6lngμ4, which is however incomplete as it is beyond the accuracy of our setup.Reuse & Permissions
  • Figure 14
    Figure 14
    Dimensionless coefficients of the expansion of the pressure in powers of g at πT=τgμ as a function of τ for Nc=3, Nf=2. The coefficients p0 and p4 are constants, p2 is quadratic in τ, while p4 shows a τ2lnτ behavior for small τ. The renormalization scale is varied through ΛMS¯=μ4μ.Reuse & Permissions
  • Figure 15
    Figure 15
    Plot of the T-dependent part of the interaction pressure δp [see Eq. (2.11)] to order g4 in the regime πTmD in units of (mDT=0)4=g¯4μ4/π4 and as a function of τ¯=T/mD. The dashed line denotes the dimensional reduction result to order g4, and the full line the HDL+ result which in this regime coincides with the order-g4 content of pIV.Reuse & Permissions
  • Figure 16
    Figure 16
    The dividing line between the regime of dimensional reduction and that of non-Fermi-liquid behavior (NFL) as given by Eq. (5.1) for Nf=3 (full lines) and Nf=2 (dashed lines), in comparison with the weak-coupling result (5.2) for the critical temperature of color superconductivity (CSC) when extrapolated to large coupling.Reuse & Permissions
  • Figure 17
    Figure 17
    The structure of the weak-coupling expansion of the interaction pressure ppSB at parametrically small T/μ as a function of the power x in Tgxμ. At Tμ, i.e. x=0, the expansion involves orders 2, 3, 4, … in g (logarithms in g are not made explicit); at Tg1/2μ where dimensional reduction overlaps with HTL/HDL resummation, the series in g involves powers 2, 3, 72, 4, 92, …; at Tgμ, where dimensional reduction ceases to be applicable, the expansion is again in even powers of g (and logs) with coefficients that at even smaller temperatures can be expanded in a series involving fractional powers of T (beginning with 2, 83, 103, 4, 143, …) and corresponding powers 2+2x, 2+83x, … of g. While subleading in the pressure, the latter contributions give the leading-order anomalous (non-Fermi-liquid) contributions to the interaction part of the entropy and specific heat at low temperature. Existing results for the various contributions are represented by full lines, as yet undetermined contributions by dashed and dashed-dotted lines. The nonperturbative barrier from the scale of magnetostatic confinement (magnetic screening mass) is indicated by the thick line marked “nonperturbative.” The region below it and up to x=1 is the regime of electrostatic QCD , while for x1 the relevant effective theory is given by nonstatic hard dense loops (HDL).Reuse & Permissions
  • Figure 18
    Figure 18
    Symbolic illustration of the analytic structure of the integrand of pringfinite on the complex plane. The unphysical pole is avoided by using the complex integration path described in the text.Reuse & Permissions
  • Figure 19
    Figure 19
    The shape of the Debye mass cutoff and the numerical integration contour chosen on the complex q0 plane, in units of r. The full line shows the deformed integration path along complex q0 while the dotted line indicates usual Minkowski space integration, extending to q0. The dashed-dotted line shows the integration path if it is restricted to stay in Minkowski space up to q0=q/2. The parameters in this and the following two figures are chosen (a) for the big circle shape (q=.5r).Reuse & Permissions
  • Figure 20
    Figure 20
    The parameters in this figure are chosen (b) for the eyeglasses shape (q=.97r).Reuse & Permissions
  • Figure 21
    Figure 21
    The parameters in this figure are chosen (c) for the two circles shape (q=1.2r).Reuse & Permissions
×

Sign up to receive regular email alerts from Physical Review D

Log In

Cancel
×

Search


Article Lookup

Paste a citation or DOI

Enter a citation
×