NLO QCD corrections to \( {\text{t}}\overline {\text{t}} {\text{b}}\overline {\text{b}} \) production at the LHC: 2. Full hadronic results
A. Bredenstein
3
A. Denner
1
S. Dittmaier
2
S. Pozzorini
0
Open Access
0
CERN Theory Division
, CERN, Case CO1600, CH-1211 GENEVA 23,
Switzerland
1
Laboratory for Particle Physics, Paul Scherrer Institut
, Wurenlingen und Villigen, CH-5232 Villigen PSI,
Switzerland
2
Physikalisches Institut, Albert-Ludwigs-Universitat Freiburg
, Hermann-Herder-Str. 3, D-79104 Freiburg,
Germany
3
Theory Center, High Energy Accelerator Research Organization (KEK)
, Tsukuba, Ibaraki 305-0801,
Japan
We present predictions for ttbb production at the LHC in next-to-leading order QCD. The precise description of this background process is a prerequisite to observe associated ttH production in the H bb decay channel and to directly measure the topquark Yukawa coupling at the LHC. The leading-order cross section is extremely sensitive to scale variations. We observe that the traditional scale choice adopted in ATLAS simulations underestimates the ttbb background by a factor two and introduce a new dynamical scale that stabilizes the perturbative predictions. We study various kinematic distributions and observe that the corrections have little impact on their shapes if standard cuts are applied. In the regime of highly boosted Higgs bosons, which offers better perspectives to observe the ttH signal, we find significant distortions of the kinematic distributions. The one-loop amplitudes are computed using process-independent algebraic manipulations of Feynman diagrams and numerical tensor reduction. We find that this approach provides very high numerical stability and CPU efficiency.
Contents
1 Introduction
Details of the calculation
2.1 Virtual corrections
2.2 Real corrections
3.1 Input parameters, jet definition, and cuts
3.2 Renormalization and factorization scales
3.3 Additional cuts
3.4 Setup I
3.7 Setup IV
A Some details on the reduction of standard matrix elements
B Benchmark numbers for the virtual corrections
Introduction
The discovery of the Higgs boson and the measurement of its interactions with massive
quarks and vector bosons represent a central goal of the ATLAS [1, 2] and CMS [3]
experiments at the Large Hadron Collider (LHC). The present limits from direct searches and
electroweak precision measurements favour a Higgs-mass range below the W-decay threshold.
In this light-Higgs scenario, the Higgs predominantly decays into bottom quarks, and its
observation is very challenging at the LHC. In the dominant Higgs-production channel, i.e. in
gluon-gluon fusion, the H bb signal is completely obscured by a huge QCD background.
Associated production mechanisms, where the Higgs boson is accompanied by a
massive gauge boson or top-quark pairs, feature more distinctive signatures that can be
exploited to reduce the background to the H bb final state. These associated
Higgsproduction channels provide direct access to the interactions of the Higgs boson with
gauge bosons and heavy quarks. Their observation would permit to test the electroweak
symmetry-breaking mechanism. But the QCD background to associated WH, ZH, and
ttH production followed by H bb decay remains a critical issue, which requires
significant progress in two directions. The low signal-to-background ratios must be increased by
means of improved selection strategies; and more precise descriptions of the background
are needed in order to reduce systematic uncertainties. In this paper we address the latter
issue by providing next-to-leading-order (NLO) QCD predictions for the irreducible QCD
background to ttH(H bb) production.
The strategies elaborated by ATLAS and CMS to identify the ttH(H bb) signal [2
9] are based on the full reconstruction of the ttbb signature, starting from a final state with
four b quarks and additional light jets. After imposing four b-taggings, a reconstruction of
the top quarks is performed. This permits to identify two b quarks as top-decay products.
The remaining two b quarks constitute a Higgs candidate, and their invariant-mass
distribution is the relevant observable to find the Higgs signal. However, the presence of multiple
b quarks and light jets in the final state represents a serious obstacle to the correct
identification of the bb Higgs candidates. Realistic simulations indicate that only about 1/3 of the
selected b-quark pairs have correct combinatorics, while the other Higgs candidates contain
b jets from top decays or miss-tagged light jets. This so-called combinatorial background
significantly dilutes the Higgs signal and increases its background contamination. The QCD
processes pp ttbb and ttjj are the main background components. The latest ATLAS and
CMS simulations [2, 3], for 30 fb1 and 60 fb1, respectively, anticipate a statistical
significance around 2 (ignoring systematic uncertainties) and a fairly low signal-to-background
ratio of order 1/10. This calls for better than 10% precision in the background description,
a very demanding requirement both from the experimental and theoretical point of vie (...truncated)