Higher-order QCD corrections to hadronic τ decays from Padé approximants

Diogo Boito, Pere Masjuan, Fabio Oliani

Research output: Contribution to journalArticleResearchpeer-review

18 Citations (Scopus)

Abstract

© 2018, The Author(s). Perturbative QCD corrections to hadronic τ decays and e+e− annihilation into hadrons below charm are obtained from the Adler function, which at present is known in the chiral limit to five-loop accuracy. Extractions of the strong coupling, αs, from these processes suffer from an ambiguity related to the treatment of unknown higher orders in the perturbative series. In this work, we exploit the method of Padé approximants and its convergence theorems to extract information about higher-order corrections to the Adler function in a systematic way. First, the method is tested in the large-β0 limit of QCD, where the perturbative series is known to all orders. We devise strategies to accelerate the convergence of the method employing renormalization scheme variations and the so-called D-log Padé approximants. The success of these strategies can be understood in terms of the analytic structure of the series in the Borel plane. We then apply the method to full QCD and obtain reliable model-independent predictions for the higher-order coefficients of the Adler function. For the six-, seven-, and eight-loop coefficients we find c5,1 = 277 ± 51, c6,1 = 3460±690, and c7,1 = (2.02±0.72)×104, respectively, with errors to be understood as lower and upper bounds. Our model-independent reconstruction of the perturbative QCD corrections to the τ hadronic width strongly favours the use of fixed-order perturbation theory (FOPT) for the renormalization-scale setting.
Original languageEnglish
Article number75
JournalJournal of High Energy Physics
Volume2018
Issue number8
DOIs
Publication statusPublished - 1 Aug 2018

Keywords

  • NLO Computations
  • QCD Phenomenology

Fingerprint

Dive into the research topics of 'Higher-order QCD corrections to hadronic τ decays from Padé approximants'. Together they form a unique fingerprint.

Cite this