No deconfinement in QCD ?

At a critical temperature QCD in the chiral limit undergoes a chiral restoration phase transition. Above the phase transition the quark condensate vanishes. The Banks-Casher relation connects the quark condensate to a density of the near-zero modes of the Dirac operator. In the Nambu-Goldstone mode the quasi-zero modes condense around zero, \lambda \rightarrow 0, and provide a nonvanishing quark condensate. The chiral restoration phase transition implies that above the critical temperature there is no any longer a condensation of the Dirac modes around zero. If a U(1)_A symmetry is also restored and a gap opens in the Dirac spectrum then the Euclidean correlation functions are SU(2N_f) \supset SU(N_f)_L \times SU(N_f)_R \times U(1)_A- symmetric. This symmetry implies that a free (deconfined) propagation of quarks in Minkowski space-time that perturbatively interact with unconfined gluons is impossible. This means that QCD above the critical temperature is not of a quark-gluon plasma origin and has a more complicated structure.