Search for charged Higgs bosons in ATLAS

Charged Higgs bosons are predicted in several extensions of the Standard Model, where the Higgs sector contains more than one doublet of complex scalars, for instance in the Minimal Supersymmetric Standard Model (MSSM). For $m_{H^+}<m_{\mathrm{top}}$, the dominant production mode for charged Higgs bosons at the LHC is via the decay $t\rightarrow bH^+$ of one of the top quarks in $t\bar{t}$ events. We present results on the search for such light charged Higgs bosons in the ATLAS experiment, with emphasis on the decay $H^+ \rightarrow \tau \nu$.


Introduction
One major task for the Large Hadron Collider (LHC) and the ATLAS experiment is the exploration of the Higgs sector in the Standard Model (SM) and beyond it. Charged Higgs bosons, H + and H − , are predicted by several extensions of the SM, such as models containing Higgs triplets and Two-Higgs-Doublet Models (2HDM) [1]. Electroweak symmetry breaking through two complex Higgs doublets lead to five physical states, out of which two are charged. The observation of charged Higgs bosons would clearly indicate physics beyond the SM. Here the type-II 2HDM is considered, which is also the Higgs sector of the Minimal Supersymmetric Standard Model (MSSM) [2]. For values of tan β (the ratio of the vacuum expectation values of the two Higgs doublets) larger than 3, typically, charged Higgs bosons decay mainly via H + → τν [3]. In the following, a branching ratio B(H + → τν) = 1 is considered.
The searches for the charged Higgs boson in tt events (see Fig. 1) with one or two light charged leptons (i.e. electrons or muons) in the final state [4] and in the case where both the τ and the W boson decay hadronically (τ+jets channel) [5] are described. 1.03 fb −1 of data from proton-proton collisions, at a centre of mass energy of √ s = 7 TeV, collected with the ATLAS experiment in 2011, are analysed. Upper limits for B(t → bH ± ) and the charged Higgs boson production in the m H + − tan β plane (m max H scenario [6]) are derived.

The τ+jets Final State
where ∆φ is the azimuthal angle between the τ-jet and the missing energy direction. The main sources of background a e-mail: daniel.pelikan@cern.ch are tt and single top-quark production, W+jets and multijet events. All these backgrounds are determined with datadriven methods.

Electron-to-τ misidentification probability with a tag-and-probe method
Background events where an electron is misidentified as a hadronically decaying τ are estimated from a control sample using a data-driven tag-and-probe method. In order to determine the misidentification probability, Z/γ * events in collision data are used. The result in data is compared to Monte Carlo (MC) and the ratio is used as a scale factor in MC.

Jet-to-τ misidentification probability from photon+jets events
In order to determine the probability for a jet to be misidentified as a hadronically decaying τ, a γ+jet control sample is used. Like jets from the hard processes in the dominant tt background, jets in the control sample originate mainly from quarks. A measurement of the probability for a jet to be misidentified as a hadronically decaying τ is performed in data and is used to predict the yield of jet-to-τ misidentification events from the most important SM backgrounds.

Multi-jet background estimate
The expectations in MC for multi-jet events have large uncertainties, for that reason this background is estimated arXiv:1201.4710v1 [hep-ex] 23 Jan 2012 EPJ Web of Conferences    from data. This is achieved by fitting its E miss T shape to data. For the shape study a control region is defined where the τ identification and the b-tagging are inverted.

Embedding method
The embedding method is used to estimate the background from true τ jets. Control samples of single-top, W+jets, and tt events with a muon in data are collected and the detector signature of this muon is replaced with that of a simulated hadronically decaying τ. Then, the reconstruction is re-applied on the new hybrid event.

Results
In Fig. 2 the m T distribution with all backgrounds after all cuts is shown. Assuming B(H + → τν) = 1, upper limits are extracted on the branching ratio B ≡ B(t → bH + ) as a function of the charged Higgs boson mass. A profile likelihood statistical analysis is performed in order to set upper limits with a 95% Confidence Level (C.L.) on the branching ratio (Fig. 3) and the H + production in the m max H scenario (Fig. 4). A limit on the branching ratio below 10% is reached.

Single-and Di-Lepton Final States
In this channel, the τ from the charged Higgs boson decays leptonically, the W boson on the other side decays either hadronically (single-lepton) or leptonically (di-lepton). To search for charged Higgs bosons in the single-and di-lepton channels, discriminating variables are used, that allow a distinction between leptons from τ → lν l ν τ (e.g. in decays of W or charged Higgs bosons) or leptons coming directly from the W boson decay. The invariant mass m bl of a bquark and a light charged lepton l (electron or muon) can be used, or more conveniently, the variable cos θ * l defined as: where p b and p l are the four-momenta of the b-quark and the lepton, m top and m W are the mass of the top-quark and the W boson.
Other discriminating variables are two new transverse mass observables [7]. In single-lepton events, the charged Higgs boson transverse mass m H T is computed as follows: (3) In the case of di-lepton events, the generalised charged Higgs boson transverse mass m H T2 is obtained as follows: where some kinematical constraints (see eq. (9) in [7]) must be fulfilled and with the definition: (5) The multi-jet background is estimated using data-driven methods, the other backgrounds are taken from simulation.

Single-Lepton Events
In order to select single-lepton tt events, the following selection cuts are applied: exactly one trigger-matched lepton with E T > 25 GeV (electron) or p T > 20 GeV (muon), at least four jets with p T > 20 GeV and |η| < 2.5, including exactly two b-jets, if |φ l,miss | < π/6.  After having assigned two light-quark jets and one b-jet to the hadronically decaying top-quark candidate, the other bjet is associated to the lepton and the charged Higgs boson transverse mass is reconstructed, see Fig. 5. To enhance the sample with leptonically decaying τs, additional cuts on cos θ * l < −0.6 and m W T < 60 GeV are applied. m W T is defined as in Eq. (1), where τ is replaced by the lepton.

Di-Lepton Events
To select di-lepton tt events the following cuts are applied: exactly two oppositely charged leptons, including at least one matched to the trigger with E T > 25 GeV (electron) or p T > 20 GeV (muon), at least two jets with p T > 20 GeV and |η| < 2.5, including exactly two b-tagged jets, for ee and µµ events, the invariant mass m ll must exceed 15 GeV and must satisfy |m ll − m Z | > 10 GeV, together with E miss T > 40 GeV, for eµ events, the scalar sum of the transverse energies of the selected leptons and jets is E T > 130 GeV.
After assignment of the two leptons and the two b-jets to their correct parents, based on cos θ * l and the distance between each lepton and b-jet, the generalised charged Higgs boson mass is reconstructed, see Fig. 6.

Limits on the Branching Ratio of t → bH +
The combined limits, from single-and di-lepton events, with a C.L. of 95% for the branching ratio B ≡ B(t → b H + ) and the limit on the production for the charged Higgs boson are shown in Fig. 7 and 8.

Conclusion
Results for the search of charged Higgs bosons in ATLAS are presented using 1.03 fb −1 of pp collision data at 7 TeV. Upper limits for the branching ratio B(t → bH + ) between 3% and 10% for charged Higgs boson masses in the range of 90 GeV < m + H < 160 GeV have been derived, with the analysis of more ATLAS data improvements are expected.