Nonexistence of the non-Gaussian fixed point predicted by the RG field theory in 4-epsilon dimensions

Abstract

The Ginzburg-Landau phase transition model is considered in d=4-epsilon dimensions within the renormalization group (RG) approach. The problem of existence of the non-Gaussian fixed point is discussed. An equation is derived from the first principles of the RG theory (under the assumption that the fixed point exists) for calculation of the correction-to-scaling term in the asymptotic expansion of the two-point correlation (Green's) function. It is demonstrated clearly that, within the framework of the standard methods (well justified in the vicinity of the fixed point) used in the perturbative RG theory, this equation leads to an unremovable contradiction with the known RG results. Thus, in its very basics, the RG field theory in 4-epsilon dimensions is contradictory. To avoid the contradiction, we conclude that such a non-Gaussian fixed point, as predicted by the RG field theory, does not exist. Our consideration does not exclude existence of a different fixed point.