Magnetic Component of Quark-Gluon Plasma is also a Liquid!

The so called magnetic scenario recently suggested in \cite{Liao_ES_mono} emphasizes the role of monopoles in strongly coupled quark-gluon plasma (sQGP) near/above the deconfinement temperature, and specifically predicts that they help reduce its viscosity by the so called "magnetic bottle" effect. Arguments for "magnetic liquid" in 1-2$T_c$ based on lattice measurement of monopole density were provided in \cite{Chernodub}. Here we present results for monopole-(anti)monopole correlation functions from the same classical molecular dynamics simulations, which are found to be in very good agreement with recent lattice results \cite{D'Alessandro:2007su}. We show that the magnetic Coulomb coupling does run in the direction $opposite$ to the electric one, as expected, and it is roughly inverse of the asymptotic freedom formula for the electric one. However, as $T$ decreases to $T_c$, the magnetic coupling never gets weak, with the plasma parameter always large enough ($\Gamma>1$). This nicely agrees with empirical evidences from RHIC experiments, implying that magnetic objects cannot have large mean free path and should also form a good liquid with low viscosity.