Renormalization and Running of Quark Mass and Field in the Regularization Invariant and MS-bar Schemes at Three and Four Loops

TL;DR

This paper derives high-order conversion formulas between MS-bar and RI schemes for quark masses and fields, highlighting the importance of higher-loop corrections for precision in lattice QCD and providing new anomalous dimensions at four loops.

Contribution

It provides explicit three- and four-loop conversion factors and anomalous dimensions, improving the precision of quark mass determinations in different renormalization schemes.

Findings

NNNLO mass conversion factor is large at 2 GeV

Higher scale (~3 GeV) improves convergence for lattice QCD

Four-loop anomalous dimensions are computed

Abstract

We derive explicit transformation formulae relating the renormalized quark mass and field as defined in the MS-bar scheme with the corresponding quantities defined in any other scheme. By analytically computing the three-loop quark propagator in the high-energy limit (that is keeping only massless terms and terms of first order in the quark mass) we find the NNNLO conversion factors transforming the MS-bar quark mass and the renormalized quark field to those defined in a ``Regularization Invariant'' (RI) scheme which is more suitable for lattice QCD calculations. The NNNLO contribution in the mass conversion factor turns out to be large and comparable to the previous NNLO contribution at a scale of 2 GeV --- the typical normalization scale employed in lattice simulations. Thus, in order to get a precise prediction for the MS-bar masses of the light quarks from lattice calculations the latter should use a somewhat higher scale of around, say, 3 GeV where the (apparent) convergence of the perturbative series for the mass conversion factor is better. We also compute two more terms in the high-energy expansion of the MS-bar renormalized quark propagator. The result is then used to discuss the uncertainty caused by the use of the high energy limit in determining the MS-bar mass of the charmed quark. As a by-product of our calculations we determine the four-loop anomalous dimensions of the quark mass and field in the Regularization Invariant scheme. Finally, we discuss some physical reasons lying behind the striking absence of zeta(4) in these computed anomalous dimensions.