Heavy-light mesons in lattice HQET and QCD

TL;DR

This paper combines HQET and relativistic QCD in lattice simulations to accurately determine the b-quark mass and Bs-meson decay constant, providing non-perturbative renormalization techniques and results consistent with experimental data.

Contribution

It introduces a finite size technique connecting small and large volume simulations for heavy quarks, and computes the renormalization group invariant b-quark mass and decay constant with high precision.

Findings

b-quark mass Mb = 6.88(10) GeV

Bs decay constant fBs = 191(6) MeV

Negligible cutoff effects confirmed by non-perturbative computations

Abstract

We present a study of a combination of HQET and relativistic QCD to extract the b-quark mass and the Bs-meson decay constant from lattice quenched simulations. We start from a small volume, where one can directly simulate the b-quark, and compute the connection to a large volume, where finite size effects are negligible, through a finite size technique. The latter consists of steps extrapolated to the continuum limit, where the b-region is reached through interpolations guided by the effective theory. With the lattice spacing given in terms of the Sommer's scale r0 and the experimental Bs and K masses, we get the final results for the renormalization group invariant mass Mb = 6.88(10) GeV, translating into mb(mb) = 4.42(6) GeV in the MSbar scheme, and fBs = 191(6) MeV for the decay constant. A renormalization condition for the chromo-magnetic operator, responsible, at leading order in the heavy quark mass expansion of HQET, for the mass splitting between the pseudoscalar and the vector channel in mesonic heavy-light bound states, is provided in terms of lattice correlations functions which well suits a non-perturbative computation involving a large range of renormalization scales and no valence quarks. The two-loop expression of the corresponding anomalous dimension in the Schroedinger functional (SF) scheme is computed starting from results in the literature; it requires a one-loop calculation in the SF scheme with a non-vanishing background field. The cutoff effects affecting the scale evolution of the renormalization factors are studied at one-loop order, and confirmed by non-perturbative quenched computations to be negligible for the numerical precision achievable at present.