Table of Contents
Fetching ...

Finite-Width Effects in Top Quark Production at Hadron Colliders

N. Kauer, D. Zeppenfeld

TL;DR

The paper develops a gauge-invariant, full-tree-level framework to include finite-width and off-shell effects in top-quark pair production with or without an additional jet. By adopting an overall-factor scheme, it blends resonant and non-resonant contributions while preserving Ward identities, enabling accurate predictions for tt and ttj backgrounds to Higgs searches at the LHC. It provides detailed matrix-element construction, phase-space partitioning, and numerical results showing how off-shell effects modify background normalizations under realistic cuts, with implications for both inclusive H→WW analyses and Weak Boson Fusion studies. The work highlights the importance of including finite-width effects for precise background modeling and lays groundwork for further refinements, such as matching to parton showers.

Abstract

Production cross sections for t\bar{t} and t\bar{t}j events at hadron colliders are calculated, including finite width effects and off resonance contributions for the entire decay chain, t --> bW --> b\ellν, for both top quarks. Resulting background rates to Higgs search at the CERN LHC are updated for inclusive H --> WW studies and for H --> ττand H --> WW decays in weak boson fusion events. Finite width effects are large, increasing t\bar{t}(j) rates by 20% or more, after typical cuts which are employed for top-background rejection.

Finite-Width Effects in Top Quark Production at Hadron Colliders

TL;DR

The paper develops a gauge-invariant, full-tree-level framework to include finite-width and off-shell effects in top-quark pair production with or without an additional jet. By adopting an overall-factor scheme, it blends resonant and non-resonant contributions while preserving Ward identities, enabling accurate predictions for tt and ttj backgrounds to Higgs searches at the LHC. It provides detailed matrix-element construction, phase-space partitioning, and numerical results showing how off-shell effects modify background normalizations under realistic cuts, with implications for both inclusive H→WW analyses and Weak Boson Fusion studies. The work highlights the importance of including finite-width effects for precise background modeling and lays groundwork for further refinements, such as matching to parton showers.

Abstract

Production cross sections for t\bar{t} and t\bar{t}j events at hadron colliders are calculated, including finite width effects and off resonance contributions for the entire decay chain, t --> bW --> b\ellν, for both top quarks. Resulting background rates to Higgs search at the CERN LHC are updated for inclusive H --> WW studies and for H --> ττand H --> WW decays in weak boson fusion events. Finite width effects are large, increasing t\bar{t}(j) rates by 20% or more, after typical cuts which are employed for top-background rejection.

Paper Structure

This paper contains 10 sections, 19 equations, 11 figures, 7 tables.

Figures (11)

  • Figure 1: Feynman diagrams contributing to $gg\to b\bar{b} e^-\bar{\nu}_e\mu^+\nu_\mu g$ in the narrow-width approximation: (a) on-shell $t\bar{t}$ production and (b) $Wt$ single top production. The double bars indicate heavy quark propagators which may be treated as on-shell particles in various approximations.
  • Figure 2: Dyson resummation of the imaginary part of the $Wb$ contribution to the top self-energy. The resummed top propagator is represented by the cross-hatched blob.
  • Figure 3: Effective $ttg$ vertex including the imaginary part of the $Wb$ loop-correction to the tree-level vertex. The imaginary part is obtained by cutting the triangle graph in all possible ways corresponding to on-shell intermediate states.
  • Figure 4: The electroweak gauge cancellation in $b W^+_L\to b W^+_L$ is perturbed by the resummed one-loop approach.
  • Figure 5: Feynman diagrams contributing to $gg\to b\bar{b} e^-\bar{\nu}_e\mu^+\nu_\mu g$ with off-shell intermediate states: double-resonant (a,b), single-resonant (c) and non-resonant (d,e,f) contributions.
  • ...and 6 more figures