No Arabic abstract
We calculate Wilson loops of various sizes up to loop order $n=20$ for lattice sizes of $L^4 (L=4, 6, 8, 12)$ using the technique of Numerical Stochastic Perturbation Theory in quenched QCD. This allows to investigate the behaviour of the perturbative series at high orders. We discuss three models to estimate the perturbative series: a renormalon inspired fit, a heuristic fit based on an assumed power-law singularity and boosted perturbation theory. We have found differences in the behavior of the perturbative series for smaller and larger Wilson loops at moderate $n$. A factorial growth of the coefficients could not be confirmed up to $n=20$. From Monte Carlo measured plaquette data and our perturbative result we estimate a value of the gluon condensate $<frac{alpha}{pi}GG>$.
We calculate perturbative Wilson loops of various sizes up to loop order $n=20$ at different lattice sizes for pure plaquette and tree-level improved Symanzik gauge theories using the technique of Numerical Stochastic Perturbation Theory. This allows us to investigate the behavior of the perturbative series at high orders. We observe differences in the behavior of perturbative coefficients as a function of the loop order. Up to $n=20$ we do not see evidence for the often assumed factorial growth of the coefficients. Based on the observed behavior we sum this series in a model with hypergeometric functions. Alternatively we estimate the series in boosted perturbation theory. Subtracting the estimated perturbative series for the average plaquette from the non-perturbative Monte Carlo result we estimate the gluon condensate.
We calculate Wilson loops of various sizes up to 20 loops in SU(3) pure lattice gauge theory at different lattice sizes for Wilson gauge action using the technique of numerical stochastic perturbation theory. This allows us to investigate the perturbative series for various Wilson loops at high loop orders. We observe differences in the behavior of those series as function of the loop order. Up to $n=20$ we do not find evidence for the factorial growth of the expansion coefficients often assumed to characterize an asymptotic series. Based on the actually observed behavior we sum the series in a model parametrized by hypergeometric functions. Alternatively we estimate the total series in boosted perturbation theory using information from the first 14 loops. We introduce generalized ratios of Wilson loops of different sizes. Together with the corresponding Wilson loops from standard Monte Carlo measurements they enable us to assess their non-perturbative parts.
The behaviour of the space-like string tension in the high temperature phase is studied. Data obtained in the $Z_2$ gauge model in (2+1) dimensions are compared with predictions of a simple model of a fluctuating flux tube with finite thickness. It is shown that in the high temperature phase contributions coming from the fluctuations of the flux tube vanish. As a consequence we also show that in (2+1) dimensional gauge theories the thickness of the flux tube coincides with the inverse of the deconfinement temperature.
We complete our high-accuracy studies of the lattice ghost propagator in Landau gauge in Numerical Stochastic Perturbation Theory up to three loops. We present a systematic strategy which allows to extract with sufficient precision the non-logarithmic parts of logarithmically divergent quantities as a function of the propagator momentum squared in the infinite-volume and $ato 0$ limits. We find accurate coincidence with the one-loop result for the ghost self-energy known from standard Lattice Perturbation Theory and improve our previous estimate for the two-loop constant contribution to the ghost self-energy in Landau gauge. Our results for the perturbative ghost propagator are compared with Monte Carlo measurements of the ghost propagator performed by the Berlin Humboldt university group which has used the exponential relation between potentials and gauge links.
We compute the low lying eigenvalues of the Hermitian Dirac operator in lattice QCD with $N_{rm f} = 2+1+1$ twisted mass fermions. We discuss whether these eigenvalues are in the $epsilon$-regime or the $p$-regime of Wilson chiral perturbation theory ($chi$PT) for twisted mass fermions. Reaching the deep $epsilon$-regime is practically unfeasible with presently typical simulation parameters, but still the few lowest eigenvalues of the employed ensemble evince some characteristic $epsilon$-regime features. With this conclusion in mind, we develop a fitting strategy to extract two low energy constants from analytical $epsilon$-regime predictions at a fixed index. Thus, we obtain results for the chiral condensate and the low energy constant $W_8$. We also discuss how to improve both the theoretical calculation and the lattice computation.