Contents
\pdfinclusioncopyfonts

=1 \addbibresourceHIGG-2017-04-PAPER.bib \AtlasTitleSearch for charged Higgs bosons decaying into top and bottom quarks at  \TeV with the ATLAS detector \AtlasAbstractA search for charged Higgs bosons heavier than the top quark and decaying via is presented. The data analysed corresponds to 36.1 fb of collisions at \TeV and was recorded with the ATLAS detector at the LHC in 2015 and 2016. The production of a charged Higgs boson in association with a top quark and a bottom quark, , is explored in the mass range from to  \GeV using multi-jet final states with one or two electrons or muons. Events are categorised according to the multiplicity of jets and how likely these are to have originated from hadronisation of a bottom quark. Multivariate techniques are used to discriminate between signal and background events. No significant excess above the background-only hypothesis is observed and exclusion limits are derived for the production cross-section times branching ratio of a charged Higgs boson as a function of its mass, which range from 2.9 pb at  \GeV to 0.070 pb at  \GeV. The results are interpreted in two benchmark scenarios of the Minimal Supersymmetric Standard Model. \AtlasRefCodeHIGG-2017-04 \PreprintIdNumberCERN-EP-2018-168 \AtlasJournalRefJHEP 11 (2018) 085 \AtlasDOI10.1007/JHEP11(2018)085 \arXivId1808.03599 \HepDataRecordhttps://www.hepdata.net/record/83203 \AtlasJournalJHEP \pdfstringdefDisableCommands\pdfstringdefDisableCommands

1 Introduction

Following the discovery of a Higgs boson, \Higgs, with a mass of around 125 \GeV and consistent with the Standard Model (SM) [HIGG-2012-27, Chatrchyan:2012ufa, HIGG-2014-14] at the Large Hadron Collider (LHC) in 2012 [Evans:2008zzb] a key question is whether this Higgs boson is the only Higgs boson, or the first observed physical state of an extended Higgs sector. No charged fundamental scalar boson exists in the SM, but many beyond the Standard Model (BSM) scenarios contain an extended Higgs sector with at least one set of charged Higgs bosons, \hplus and \hminus, in particular two-Higgs-doublet models (2HDM) [Lee:1973iz, Akeroyd:2016ymd, Gunion:2002zf, Branco:2011iw] and models containing Higgs triplets [trip1, trip2, trip3, trip4, trip5].

The production mechanisms and decay modes of a charged Higgs boson111For simplicity in the following, charged Higgs bosons are denoted \hplus, with the charge-conjugate \hminus always implied. Similarly, the difference between quarks and antiquarks, and , is generally understood from the context, so that e.g. means both and . depend on its mass, \mhp. This analysis searches for heavy charged Higgs bosons with , where and are the masses of the top and bottom quarks, respectively. The dominant production mode is expected to be in association with a top quark and a bottom quark (\tbhplus), as illustrated in Figure 1.

Figure 1: Leading-order Feynman diagram for the production of a heavy charged Higgs boson () in association with a top quark and a bottom quark (\tbhplus).

In the 2HDM, \hplus production and decay at tree level depend on its mass and two parameters: \tanbeta and , which are the ratio of the vacuum expectation values of the two Higgs doublets and the mixing angle between the CP-even Higgs bosons, respectively. The dominant decay mode for heavy charged Higgs bosons is in a broad range of models [Heinemeyer:2013tqa, deFlorian:2016spz]. In particular, this is the preferred decay mode in both the decoupling limit scenario and the alignment limit , where the lightest CP-even neutral Higgs boson of the extended Higgs sector has properties similar to those of the SM Higgs boson [Gunion:2002zf]. For lower \mhp, the dominant decay mode is . It is also predicted that this decay mode becomes more relevant as the value of \tanbeta increases, irrespective of \mhp. Therefore, the and decays naturally complement each other in searches for charged Higgs bosons.

Limits on charged Higgs boson production have been obtained by many experiments, such as the LEP experiments with upper limits on \hplus production in the mass range 40–100 \GeV [Abbiendi:2013hk], and CDF and DØ at the Tevatron that set upper limits on the branching ratio for  [Aaltonen:2009ke, Abazov:2009aa]. The CMS Collaboration has performed direct searches for heavy charged Higgs bosons in 8 \TeV proton–proton () collisions. By assuming the branching ratio , an upper limit of 2.0–0.13 pb was obtained for the production cross-section for  [Khachatryan:2015qxa]. The ATLAS Collaboration has searched for similar heavy charged Higgs boson production in the decay channel at 8 \TeV, setting upper limits on the production cross-section times the branching ratio of 6–0.2 pb for  [Aad:2015typ]. Indirect constraints can be obtained from the measurement of flavour-physics observables sensitive to charged Higgs boson exchange. Such observables include the relative branching ratios of or meson decays, meson mixing parameters, the ratio of the decay partial widths , as well as the measurements of decays [Deschamps:2009rh, Arbey:2017gmh]. The relative branching ratio , where denotes or , are especially sensitive to contributions from new physics. Measurements from BaBar [Lees:2013uzd] exclude \hplus for all \mhp and \tanbeta values in a Type-II 2HDM. However, more recent measurements from Belle [Hirose:2017dxl, Sato:2016svk, Huschle:2015rga] and LHCb [Aaij:2017deq] place a weaker constraint on the allowed range of values. A global fit combining the most recent flavour-physics results [Arbey:2017gmh] sets a lower limit at 95% confidence level on the charged Higgs boson mass of for and for lower \tanbeta values, assuming a Type-II 2HDM.

This paper presents a search for \hplus production in the \hplustb decay mode using collisions at  \TeV. Events with one charged lepton () and jets in the final state (\ljets final state) and events with two charged leptons and jets in the final state (\dilepton final state) are considered. Exclusive regions are defined according to the number of jets and those that are tagged as originating from the hadronisation of a -quark. In order to separate the signal from the SM background, multivariate discriminants are employed in the regions where the signal contributions are expected to be largest. Limits on the \hplustb production cross-section are set by means of a simultaneous fit of binned distributions of multivariate discriminants in the signal-rich regions and inclusive event yields in the signal-depleted regions. The results are interpreted in two benchmark scenarios of the Minimal Supersymmetric Standard Model (MSSM): the \mhmodm scenario [Carena:2013ytb] and the hMSSM [Djouadi:2013uqa]. Both scenarios exploit the MSSM in such a way that the light CP-even Higgs boson can be interpreted as the observed Higgs boson with  \GeV. Limits on the value of \tanbeta are extracted as a function of the charged Higgs boson mass. Finally, the excluded range of \mhp and \tanbeta values from the and  [ATLAS-HIGG-2016-11] searches at  \TeV are superimposed, providing a summary of the ATLAS sensitivity to \hplus through the two decay modes.

The paper is organised as follows. Section 2 briefly describes the ATLAS detector. The samples of simulated events used for the analysis are summarised in Section 3. Section 4 presents the reconstruction of objects in ATLAS and the event selection. Section 5 describes the analysis strategy while systematic uncertainties are discussed in Section 6. The statistical analysis of the data is described in Section 7 and the results are presented in Section 8. Finally, a summary is given in Section 9.

2 ATLAS detector

The ATLAS detector [PERF-2007-01] at the LHC is a multipurpose particle detector with a forward–backward symmetric cylindrical geometry and near coverage around the collision point.222 ATLAS uses a right-handed coordinate system with its origin at the nominal interaction point (IP) in the centre of the detector and the -axis along the beam pipe. The -axis points from the IP to the centre of the LHC ring, and the -axis points upwards. Cylindrical coordinates are used in the transverse plane, being the azimuthal angle around the -axis. The pseudorapidity is defined in terms of the polar angle as . Angular distance is measured in units of (pseudorapidity and azimuthal angle). Alternatively, the distance is used, where is the rapidity of a particle of energy and momentum component along the beam axis. The ATLAS detector consists of an inner tracking detector (ID) surrounded by a thin superconducting solenoid producing a 2 T axial magnetic field, electromagnetic (EM) and hadronic calorimeters, and an external muon spectrometer (MS) incorporating three large toroid magnet assemblies. The ID contains a high-granularity silicon pixel detector, including an insertable B-layer [IBL] added in 2014 as a new innermost layer, and a silicon microstrip tracker, providing precision tracking in the pseudorapidity range . The silicon detectors are complemented by a transition radiation tracker providing tracking and electron identification information for . The EM sampling calorimeter uses lead as the absorber material and liquid argon (LAr) as the active medium, and is divided into barrel () and endcap () regions. Hadron calorimetry is also based on the sampling technique, with scintillator tiles or LAr as the active medium, and with steel, copper, or tungsten as the absorber material. The calorimeters cover . The MS measures the deflection of muons with using multiple layers of high-precision tracking chambers located in a toroidal field in the central and endcap regions of ATLAS. The field integral of the toroids ranges between and across most of the detector. The MS is also instrumented with separate trigger chambers covering . A two-level trigger system, with the first level implemented in custom hardware and followed by a software-based second level, is used to reduce the trigger rate to around 1 kHz for offline storage [trigger2015].

3 Signal and background modelling

The \tbhplus process was modelled with \MGMCatNLO (\mg5_\amc[Alwall:2014hca] at next-to-leading order (NLO) in QCD [Degrande:2015vpa] using a four-flavour scheme (4FS) implementation with the \nnpdf2.3NLO [Ball:2012cx] parton distribution function (PDF).333Five-flavour scheme (5FS) PDFs consider -quarks as a source of incoming partons and the -quarks are therefore assumed to be massless. In contrast, 4FS PDFs only include lighter quarks and gluons, allowing the -quark mass to be taken into account properly in the matrix element calculation. Parton showering and hadronisation were modelled by \PYTHIAV8.186 [Sjostrand:2007gs] with the A14 [a14] set of underlying-event (UE) related parameters tuned to ATLAS data (tune). For the simulation of the \tbhplus process, the narrow-width approximation was used. This assumption has a negligible impact on the analysis for the models considered in this paper, as the experimental resolution is much larger than the \hplus natural width. Interference with the SM background is neglected.

Altogether 18 \hplus mass hypotheses are used, with 25 \GeV mass steps between an \hplus mass of 200 \GeV and 300 \GeV, 50 \GeV steps between 300 \GeV and 400 \GeV, 100 \GeV steps between 400 \GeV and 1000 \GeV and 200 \GeV steps from 1000 \GeV to 2000 \GeV. The step sizes are selected to match the expected resolution of the \hplus signal. The samples were processed with a fast simulation of the ATLAS detector [fastsim]. Unless otherwise indicated, the cross-section of the signal is set to 1 pb, for easy rescaling to various model predictions. Only the \hplus decay into is considered, and the top quark decays according to the SM predictions.

The nominal sample used to model the \ttbar background was generated using the \POWHEGBOXVv2 NLO-in-QCD generator [Nason:2004rx, Frixione:2007vw, Alioli:2010xd, Campbell:2014kua], referred to as \powheg in the remainder of this article, with the \nnpdf3.0NLO PDF set [Ball:2014uwa]. The \hdamp parameter, which controls the transverse momentum \pt of the first additional emission beyond the Born configuration, was set to 1.5 times the top quark mass [ATL-PHYS-PUB-2016-020]. Parton shower and hadronisation were modelled by \PYTHIAV8.210 [Sjostrand:2014zea] with the A14 UE tune. The sample was normalised to the \toppp2.0 [Czakon:2011xx] theoretical cross-section of  pb, calculated at next-to-next-to-leading order (NNLO) in QCD including resummation of next-to-next-to-leading logarithmic (NNLL) soft gluon terms [ref:xs1, ref:xs2, ref:xs3, ref:xs4, ref:xs5]. The generation of the \ttbar sample was performed inclusively, with all possible flavours of additional jets produced. The decay of - and -hadrons was simulated with the \evtgen v1.2.0 [Lange:2001uf] program. The \ttbar+ jets background is categorised according to the flavour of additional jets in the event, using the same procedure as described in Ref. [HIGG-2013-27]. The \ttbar+ additional heavy-flavour (HF) jets background is subdivided into the categories \ttgeb and \ttgec, depending on whether the additional HF jets originate from hadrons containing - or -quarks. Particle jets were reconstructed from stable particles (mean lifetime  seconds) at generator level using the anti-\kt algorithm [Cacciari:2008gp] with a radius parameter of 0.4, and were required to have and . If at least one particle-level jet in the event is matched () to a -hadron (not originating from a -decay) with  \GeV, the event is categorised as \ttgeb. In the remaining events, if at least one jet is matched to a -hadron (not originating from a  decay) but no -hadron, the event is categorised as \ttgec. Events with \ttbar+ jets that belong to neither the \ttgeb nor \ttgec category are called \ttbar+ light events.

For the \ttbin process, subcategories are defined in accord with the matching between particle-level jets and the -hadrons not from -decay: events where exactly two jets are matched to -hadrons (\ttbb), events where exactly one jet is matched to a -hadron (\ttb), events where exactly one jet is matched to two or more -hadrons (\ttB), and all other events (\ttbbb). Events where the additional HF jets can only be matched to -hadrons from multi-parton interactions and final-state gluon radiation are considered separately and labelled as \ttb (MPI/FSR).

To model the irreducible \ttbin background to the highest available precision, the \ttbin events from the nominal \powheg+\pythia8 simulation are reweighted to an NLO prediction of including parton showering and hadronisation from \SHERPAV2.1.1 [Gleisberg:2008ta, Cascioli:2013era] with \ol [Cascioli:2011va]. This sample was generated using the 4FS PDF set CT10F4 [Lai:2010vv]. The renormalisation scale (\muR) for this sample was set to the  [Cascioli:2013era, MorettiPhD], and the factorisation (\muF) and resummation (\muQ) scales to . A first type of reweighting is performed in the \ttgeb subcategories, using a method similar to the one outlined in Ref. [Aaboud:2017rss]. The reweighting corrects the relative normalisation of the \ttgeb subcategories to match the predictions from \SHERPA, while keeping the overall \ttgeb normalisation unchanged. After applying the first reweighting based on the relative normalisation of the \ttgeb subcategories, a second type of reweighting is derived and performed on several kinematic variables sequentially. First the \pt of the \ttbar system is reweighted, and secondly the \pt of the top quarks. The final reweighting is performed depending on the type of \ttgeb events. If there is only one additional HF jet, the \pt of that jet is used in the final reweighting. If there is more than one additional HF jet, first the between the HF jets is reweighted and then the \pt of the HF dijet system. A closure test is performed on each of the reweighted kinematic variables, showing a reasonable level of agreement between the reweighted \powheg+\pythia8 sample and the \SHERPAsample.

The \POWHEGBOXVv1 generator was used to produce the samples of single-top-quark backgrounds, with the CT10 PDF set. Overlaps between the \ttbar and final states were handled using the ‘diagram removal’ scheme [Frixione:2008yi]. The -channel single-top-quark events were generated using the \POWHEGBOXVv1 generator with the 4FS for the NLO matrix element calculations and the fixed 4FS PDF set CT10F4. The top quarks were decayed with \madspin [Artoisenet:2012st], which preserves the spin correlations. The samples were interfaced to \PYTHIAV6.428 [Sjostrand:2006za] with the \perugia 2012 UE tune [Skands:2010ak]. The single-top-quark and -channel samples were normalised to the approximate NNLO (aNNLO) theoretical cross-section [Kidonakis:2010ux, Kidonakis:2010tc, Kidonakis:2011wy].

Samples of +jets events were generated using \SHERPAV2.2.1 [Gleisberg:2008ta]. Matrix elements were calculated for up to 2 partons at NLO and 4 partons at LO using \comix [Gleisberg:2008fv] and \ol and merged with the \SHERPAparton shower [Schumann:2007mg] using the ME+PS@NLO prescription [Hoeche:2012yf]. The NNPDF3.0NNLO PDF set was used together with a dedicated parton shower tune developed by the \SHERPAauthors. The +jets events were normalised to the NNLO cross-sections [Melnikov:2006kv, Gavin:2010az, Li:2012wna, Bardin:2012jk, Arbuzov:2012dx].

Samples of \ttV () events were generated at NLO in the matrix elements calculation using \mg5_\amc with the \nnpdf3.0NLO PDF set interfaced to \PYTHIAV8.210 with the A14 UE tune. The \ttH process was modelled using \mg5_\amc with NLO matrix elements, \nnpdf3.0NLO PDF set and factorisation and renormalisation scales set to , where \mtrans is defined as the scalar sum of the transverse masses of all final-state particles. The events were interfaced to \PYTHIAV8.210 with the A14 UE tune. Variations in \ttH production due to the extended Higgs sector are not considered in this analysis, since the contribution from the \ttH background is found to be small. Measurements of the \ttH production cross-section are compatible with the SM expectation [Aaboud:2018urx, Sirunyan:2018hoz].

The minor backgrounds, consisting of the production of a single top quark in association with a Higgs boson and jets (\tHjb), and the production of a single top quark, a boson and a Higgs boson (\WtH), are treated as one background. The \tHjb process was simulated with \mg5_\amc interfaced to \PYTHIAV8.210 and the CT10 PDF set, and \WtH was modelled with \mg5_\amc interfaced to \HERWIGpp [Bahr:2008pv] using the CTEQ6L1 PDF set [cteq6l1]. Additional minor SM backgrounds (diboson production, single top -channel, , , , ) were also simulated and accounted for, even though they contribute less than 1% in any analysis region.

Except where otherwise stated, all simulated event samples were produced using the full ATLAS detector simulation [Aad:2010ah] based on \GEANT[Agostinelli:2002hh]. Additional pile-up interactions were simulated with \PYTHIAV8.186 using the A2 set of tuned parameters [ATLAS:2012uec] and the MSTW2008LO PDF set [Martin:2009iq], and overlaid onto the simulated hard-scatter event. All simulated samples were reweighted such that the average number of interactions per bunch crossing (pile-up) matches that of the data. In the simulation, the top quark mass was set to . Decays of - and -hadrons were performed by \evtgen v1.2.0, except in samples simulated by the \SHERPAevent generator.

The samples and their basic generation parameters are summarised in Table 1.


Physics process
Generator Parton shower Cross-section PDF set Tune
generator normalisation
\tbhplus \mg5_\amc \PYTHIAV8.186 \nnpdf2.3NLO A14
\ttbar+ jets \POWHEGBOXVv2 \PYTHIAV8.210 NNLO+NNLL \nnpdf3.0NLO A14
\SHERPAV2.1.1 \SHERPAV2.1.1 NLO for CT10F4 \SHERPAdefault
\ttV \mg5_\amc \PYTHIAV8.210 NLO \nnpdf3.0 A14
\ttH \mg5_\amc \PYTHIAV8.210 NLO \nnpdf3.0NLO A14
Single top, \POWHEGBOXVv1 \PYTHIAV6.428 aNNLO CT10 Perugia 2012
Single top, -channel \POWHEGBOXVv1 \PYTHIAV6.428 aNNLO CT10F4 Perugia 2012
+jets \SHERPAV2.2.1 \SHERPAV2.2.1 NNLO \nnpdf3.0NNLO \SHERPAdefault
+jets \SHERPAV2.2.1 \SHERPAV2.2.1 NNLO \nnpdf3.0NNLO \SHERPAdefault
Table 1: Nominal simulated signal and background event samples. The generator, parton shower generator and cross-section used for normalisation are shown together with the applied PDF set and tune. The event sample generated using \SHERPAV2.1.1 is used to reweight the events from the \ttbin process in the \ttbar+ jets sample.

4 Object and event selection

The data used in this analysis were recorded in 2015 and 2016 from  \TeV  collisions with an integrated luminosity of \lumi. Only runs with stable colliding beams and in which all relevant detector components were functional are used. Events are required to have at least one reconstructed vertex with two or more tracks with  \GeV. The vertex with the largest sum of the squared \pt of associated tracks is taken as the primary vertex.

Events were recorded using single-lepton triggers, in both the \ljets and \dilepton final states. To maximise the event selection efficiency, multiple triggers were used, with either low \pt thresholds and lepton identification and isolation requirements, or with higher \pt thresholds but looser identification criteria and no isolation requirements. Slightly different sets of triggers were used for 2015 and 2016 data. For muons, the lowest \pt threshold was 20 (26) \GeV in 2015 (2016), while for electrons, triggers with a \pt threshold of 24 (26) \GeV were used. Simulated events were also required to satisfy the trigger criteria.

Electrons are reconstructed from energy clusters in the EM calorimeter associated with tracks reconstructed in the ID [ATLAS-CONF-2016-024]. Candidates in the calorimeter transition region are excluded. Electrons are required to satisfy the tight identification criterion described in Ref. [ATLAS-CONF-2016-024], based on shower-shape and track-matching variables. Muons are reconstructed from track segments in the MS that are matched to tracks in the ID [Aad:2016jkr]. Tracks are then re-fit using information from both detector systems. The medium identification criterion described in Ref. [Aad:2016jkr] is used to select muons. To reduce the contribution of leptons from hadronic decays (non-prompt leptons), both the electrons and muons must satisfy isolation criteria. These criteria include both track and calorimeter information, and have an efficiency of 90% for leptons with a \pt of 25 \GeV, rising to 99% above 60 \GeV, as measured in  [ATLAS-CONF-2016-024] and  [Aad:2016jkr] samples. Finally, the lepton tracks must point to the primary vertex of the event: the longitudinal impact parameter must satisfy  mm, while the transverse impact parameter significance must satisfy, (3) for electrons (muons).

Jets are reconstructed from three-dimensional topological energy clusters [PERF-2014-07] in the calorimeter using the anti-\kt jet algorithm [Cacciari:2008gp, Fastjet] with a radius parameter of 0.4. Each topological cluster is calibrated to the EM scale response prior to jet reconstruction. The reconstructed jets are then calibrated to the jet energy scale (JES) derived from simulation and in situ corrections based on  \TeV data [PERF-2016-04]. After energy calibration, jets are required to have and . Quality criteria are imposed to identify jets arising from non-collision sources or detector noise, and events containing any such jets are removed [Run2-jet-cleaning]. Finally, to reduce the effect of pile-up an additional requirement using information about the tracks and the primary vertex associated to a jet (Jet Vertex Tagger) [Aad:2015ina] is applied for jets with  \GeV and .

Jets are identified as containing the decay of a -hadron (-tagged) via an algorithm using multivariate techniques to combine information from the impact parameters of displaced tracks with the topological properties of secondary and tertiary decay vertices reconstructed within the jet [Aad:2015ydr, ATL-PHYS-PUB-2016-012]. Jets are -tagged by directly requiring the output discriminant of the -tagging algorithm to be above a threshold. A criterion with an efficiency of 70% for -jets in \ttbar events is used to determine the -jet multiplicity for all final states and \hplus masses. For this working point, the -jet and light-jet rejection factors are 12 and 381, respectively. For  \GeV, five exclusive efficiency bins are defined using the same -tagging discriminant: 0–60%, 60–70%, 70–77%, 77–85% and 85–100%, following the procedure described in Ref. [HIGG-2013-23]. These step-wise efficiencies are used as input to the kinematic discriminant described in Section 5. When ‘a -tagged jet’ is mentioned without any further specification, an efficiency of 70% is implied.

To avoid counting a single detector response as two objects, an overlap removal procedure is used. First, the closest jet within of a selected electron is removed. If the nearest jet surviving this selection is within of the electron, the electron is discarded, to ensure it is sufficiently separated from nearby jet activity. Muons are removed if they are separated from the nearest jet by , to reduce the background from muons from HF decays inside jets. However, if this jet has fewer than three associated tracks, the muon is kept and the jet is removed instead; this avoids an inefficiency for high-energy muons undergoing significant energy loss in the calorimeter.

The missing transverse momentum in the event is defined as the negative vector sum of the \pt of all the selected electrons, muons and jets described above, with an extra term added to account for energy in the event that is not associated with any of these. This extra term, referred to as the ‘soft term’ in the following, is calculated from ID tracks matched to the primary vertex to make it resilient to pile-up contamination [PERF-2014-04, PERF-2016-07, ATLAS-CONF-2018-023]. The missing transverse momentum is not used for event selection but is an input to the multivariate discriminants.

Events are required to have at least one electron or muon. The leading lepton must be matched to a lepton with the same flavour reconstructed by the trigger algorithm within , and have a  \GeV. Additional leptons are required to have  \GeV, or  \GeV for events with two electrons. The latter requirement reduces the bakground due to jets and photons that are misidentified as electrons. Events in the \ljets channel and the \dilepton channel are required to be mutually exclusive. Electrons or muons from decays are also included in the analysis.

For the \ljets channel, five or more jets, of which at least two jets have to be -tagged, are required. For the \dilepton channel, events with two leptons with opposite charge are selected, and at least three jets are required, of which two or more must be -tagged. In the and channels, the dilepton invariant mass must be  \GeV and outside the boson mass window of 83–99 \GeV.

5 Analysis strategy

After the event selection, the samples in both the \dilepton and the \ljets final states contain mostly \ttbar events. Events passing the event selection are categorised into separate regions according to the number of reconstructed jets and -tagged jets. The regions where is enhanced relative to the backgrounds are referred to as signal regions (SRs), whereas the remaining regions are referred to as control regions (CRs).

For the \ljets final state, two CRs (\FiveJexTwoBex and \SixJinTwoBex)444 means that jets are found in the event, and among them are -tagged. and four SRs (\FiveJexThreeBex, \FiveJexFourBin, \SixJinThreeBex and \SixJinFourBin) are defined, while in the \dilepton final state, two CRs (\ThreeJexTwoBex and \FourJinTwoBex) and two SRs (\FourJinThreeBex and \FourJinFourBin) are defined for all mass hypotheses. In addition, for the \dilepton final state, the region with three -tagged jets and no other jets (\ThreeJexThreeBex) is considered a SR for and a CR for due to the change in expected signal yield for the different \hplus mass hypotheses.

In the SRs, for each \hplus mass hypothesis a different discriminating variable based on boosted decision trees (BDTs) is defined. In order to separate the \hplus signal from the SM background, the binned output of this variable is used together with the total event yields in the CRs in a combined profile likelihood fit. The fit simultaneously determines both the signal and background yields, while constraining the overall background model within the assigned systematic uncertainties. The event yields in the CRs are used to constrain the background normalisation and systematic uncertainties. In the following subsections the background estimate and the design of the multivariate discriminator are described. The profile likelihood fit, including the treatment of backgrounds in the fit, is described in detail in Section 7.

5.1 Background estimate

The background from processes with prompt leptons is estimated using the simulated event samples described in Section 3. For \ttbar production, the number of events with high leading jet \pt is overestimated in the simulation, and a reweighting function for the leading jet \pt distribution is determined by comparing simulation with data in a \ljets CR that requires exactly four jets and at least two -tagged jets. This function is validated in the dilepton channel and applied to both channels.

The normalisation of the +HF jets backgrounds is corrected by a factor of , extracted from dedicated control regions in data, defined by requiring two opposite-charge same-flavour leptons ( or ) with an invariant mass compatible with the boson mass, .

Processes that do not contain enough prompt electrons or muons from or boson decays can still satisfy the selection criteria if they contain non-prompt leptons. The leading sources of non-prompt leptons in the \ljets final state are from semileptonic hadron decays or misidentified jets in multi-jet production. In the \dilepton final state, the dominant source of non-prompt leptons is from misidentified jets as leptons arising from +jets or \ljets  production. These backgrounds are estimated using data. For the \ljets final state a matrix method [TopFakes8TeV] is employed. An event sample that is enriched in non-prompt leptons is selected by using looser isolation or identification requirements for the lepton. These events are then weighted according to the efficiencies for both the prompt and non-prompt leptons to pass the tighter default selection. These efficiencies are measured using data in dedicated CRs. In the \dilepton final state, this background is estimated from simulations, and the normalisation is determined by comparing data and simulations in a CR of same-sign dilepton events. The contribution of multi-jet events to the \dilepton final state is found to be negligible.

The expected event yields of all SM processes and the number of events observed in the data are shown in Figure 2 for the \dilepton and the \ljets final states before performing the fit to data. The expected \hplus signal yields for , assuming a cross-section times branching ratio of 1 pb, are also shown.

Figure 2: Comparison of predicted and observed event yields. Each background process is normalised according to its cross-section and the prediction has not been fitted to the data. The includes contributions from \ttW, \ttZ and \ttH. A signal with , normalised to a cross-section times branching ratio for \hplustb of 1 pb, is shown as a dashed line. The lower panel displays the ratio of the data to the total prediction. The hatched bands show uncertainties before the fit to the data, which are dominated by systematic uncertainties as discussed in Section 6. The comparison is shown for all signal and control regions used in the analysis. For the \dilepton final state: CR \ThreeJexTwoBex, CR/SR \ThreeJexThreeBex, CR \FourJinTwoBex, SR \FourJinThreeBex, SR \FourJinFourBin. For the \ljets final state: CR \FiveJexTwoBex, SR \FiveJexThreeBex, SR \FiveJexFourBin, CR \SixJinTwoBex, SR \SixJinThreeBex, SR \SixJinFourBin.

5.2 Multivariate analysis

The training of the BDTs that are used to discriminate signal from background in the SRs is performed with the TMVA toolkit [tmva]. BDTs are trained separately for each value of the 18 generated \hplus masses and for each SR against all the backgrounds (\ljets channel) or the \ttbar background (\dilepton channel). For the BDT training in the \ljets channel, the SRs \FiveJexThreeBex and \FiveJexFourBin are treated as one region, in order to increase the number of simulated events available for training.

The BDT variables include various kinematic quantities with the optimal discrimination against the \ttbin background. For \hplus masses above 400 \GeV the most important variables in the \ljets final state are the scalar sum of the \pt of all jets, \Htjets, and the leading jet \pt. For a mass at or below 300 \GeV, a kinematic discriminant, , as described below, is used as an input variable for the BDT. The kinematic discriminant, , and the invariant mass of the pair of jets that are not -tagged and have the smallest are the most important variables in the low mass range. The latter variable is not used in the \FiveJexFourBin SR, where it is not well defined.

The kinematic discriminant, , is a variable reflecting the probability that an event is compatible with the and the hypotheses, and is defined as , where and are probability density functions for under the signal hypothesis and background (\ttbar) hypothesis, respectively. Here, the event variable indicates the set of the missing transverse momentum and the four-momenta of reconstructed electrons, muons and jets.

The probability is defined as the product of the probability density functions for each of the reconstructed invariant masses in the event:

  • the mass of the semileptonically decaying top quark, ,

  • the mass of the hadronically decaying boson, ,

  • the difference between the masses of the hadronically decaying top quark and the hadronically decaying boson , and

  • the difference between the mass of the charged Higgs boson and the mass of the leptonically or hadronically decaying top quark, or , depending on whether the top quark from the charged Higgs boson decays leptonically or hadronically.

In this context or refer to the quarks from the boson decay, and to the lepton and neutrino from the other boson decay, to the -quark from the hadronic top quark decay, to the -quark from the leptonic top quark decay and to the -quark directly from the \hplus decay. The probability is constructed from probability density functions obtained from simulated \ttbar events. For the SRs with five jets, is defined using the same invariant masses as above. The jet that does not originate from a top quark decay is used instead of . For the SRs with at least six jets the power of the discriminant is improved by using the invariant mass of the two highest-\pt jets not originating from the hadronisation of , , or instead of or .

The functional form of the probability density functions is obtained from simulation using the reconstructed masses of jets and leptons matched to simulated partons and leptons for and \ttbar. The neutrino four-momentum is derived with the assumption that the missing transverse momentum is solely due to the neutrino; the constraint is used to obtain . If two real solutions exist, they are sorted according to the absolute value of their , i.e., . In approximately 60% of the cases is closer than to the generator-level neutrino . Two different probability density functions are constructed, one for each solution, and the probability is defined as a weighted average of the two probability density functions. The weight is taken as the fraction of the corresponding solution being closer to the generated neutrino . Also, if no real solution exists, the and components are scaled by a common factor until the discriminant of the quadratic equation is exactly zero, yielding only one solution.

When evaluating and for the calculation of , all possible parton–jet assignments are considered since the partonic origin of the jets is not known. In order to suppress the impact from parton–jet assignments that are inconsistent with the correct parton flavours, a weighted average over all parton-jet assignments is used. The value of and for each parton–jet assignment is weighted with a probability based on the -tagging discriminant value of each jet. The distribution of the step-wise efficiencies of the -tagging algorithm, as described in Section 4, is used as a probability density function, with the -jet hypothesis for generated -quarks and the light-jet hypothesis for other generated partons. Due to the large number of events in which and cannot be matched to different jets, the average of two different probability density functions, where either all partons can be matched to jets or only one jet can be matched to and , is used. This discriminant gives better background suppression than would be obtained by adding the kinematic input variables directly to the BDT.

In the \dilepton final state, approximately ten optimal kinematic variables from the analysis objects and their combinations were selected for each SR, independently for the low-mass region ( \GeV) and the high-mass region ( \GeV). For the high-mass region, the most important variables are the scalar sum of the \pt of all jets and leptons, \Htall, and the transverse momentum of the jet pair with maximum \pt. For the low-mass region, the smallest invariant mass formed by two -tagged jets and the smallest invariant mass formed by a lepton and a -tagged jet, are among the most important variables.

All BDT input variables in the \ljets and \dilepton final states are listed in the Appendix. In most regions, the distributions show a reasonable level of agreement between simulation and data within the systematic and statistical uncertainties before the fit to the data (pre-fit). As examples, Figures 3 and 4 show the distribution of the observed and pre-fit expected event yields for \Htjets in the \ljets channel and \Htall in the \dilepton channel. Figure 5 shows the expected BDT output distributions, normalised to unity, for selected \hplus signal samples and the background processes in the SRs.

[] []

[] []

Figure 3: Distributions of the \Htjets variable before the fit to the data in the four SRs of the \ljets channel: (a) \FiveJexThreeBex, (b) \SixJinThreeBex, (c) \FiveJexFourBin, (d) \SixJinFourBin. Each background process is normalised according to its cross-section and the normalisation of the \ttbin and \ttcin backgrounds corresponds to the prediction from \powheg+\pythia8 for the fraction of each of these components relative to the total prediction. The includes contributions from \ttW, \ttZ and \ttH. In addition, the expectation for a 200 \GeV signal is shown for a cross-section times branching ratio of 1 pb. The lower panels display the ratio of the data to the total prediction. The hatched bands show the pre-fit uncertainties. The level of agreement is improved post-fit due to the adjustment of the normalisation of the \ttgeb and \ttgec backgrounds and the other nuisance parameters by the fit.

[] []

[]

Figure 4: Distributions of the \Htall variable before the fit to the data in the three SRs of the \dilepton channel: (a) \ThreeJexThreeBex, (b) \FourJinThreeBex and (c) \FourJinFourBin. Each background process is normalised according to its cross-section and the normalisation of the \ttbin and \ttcin backgrounds corresponds to the prediction from \powheg+\pythia8 for the fraction of each of these components relative to the total prediction. The includes contributions from \ttW, \ttZ and \ttH. In addition, the expectation for a 200 \GeV signal is shown for a cross-section times branching ratio of 1 pb. The lower panels display the ratio of the data to the total prediction. The hatched bands show the pre-fit uncertainties. The level of agreement is improved post-fit due to the adjustment of the normalisation of the \ttgeb and \ttgec backgrounds and the other nuisance parameters by the fit.

[] []

[] []

[] []

Figure 5: The expected output distributions of the BDTs employed for \hplus masses of 200 \GeV and 800 \GeV for SM backgrounds and \hplus signal in the three \ljets and the three \dilepton SRs used in the BDT training: (a) \ljets final state, \FiveJexThreeBin, (b) \ljets final state, \SixJinThreeBex, (c) \ljets final state, \SixJinFourBin, (d) \dilepton final state, \ThreeJexThreeBex, (e) \dilepton final state, \FourJinThreeBex and (f) \dilepton final state, \FourJinFourBin. All distributions are normalised to unity.

6 Systematic uncertainties

Systematic uncertainties from various sources affect this search, such as uncertainties in the luminosity measurement, the reconstruction and calibration of physics objects, in particular -tagged jets, and the modelling of the signal and background processes. Uncertainties can either modify the normalisation of the signal and background processes, change the shape of the final distributions, or both. The experimental uncertainties were obtained from dedicated analyses detailed in the corresponding references. The uncertainties related to this analysis are described in this section. For a precise treatment, the uncertainties are split into several components as explained in the following. The exact number of components for each category is listed in Table 2. The most important uncertainties are related to jet flavour tagging, background modelling, jet energy scale and resolution and the limited number of events in the simulation samples. The impact of all systematic uncertainties is listed in Table 5 in Section 8.

Systematic uncertainty Type Number of components
Luminosity N 1
Pile-up NS 1
Electron reconstruction NS 6
Muon reconstruction NS 13
Jet and \METreconstruction NS 28
Flavour tagging, 70% efficiency calibration (*) NS 27
Flavour tagging, step-wise efficiency calibration (*) NS 126
Signal QCD scale and PDF NS 31
Background modelling, \ttbar+ jets NS 29
Background modelling, other top NS 25
Background modelling, non-top (\ljets final state) N 13
Background modelling, non-top (\dilepton final state) N 4
Table 2: List of systematic uncertainties considered. The details of the systematic uncertainties are described in Section 6. ‘N’ indicates that the uncertainty is taken as normalisation-only for all processes and channels affected, while ‘NS’ means that the uncertainty applies to both normalisation and shape. The systematic uncertainties are split into several components for a more accurate treatment.
Flavour-tagging uncertainties marked (*) are different for the two sets of calibrations: the step-wise efficiency calibration for  \GeV, and the 70% efficiency point calibration elsewhere.

The combined uncertainty in the integrated luminosity for the data collected in 2015 and 2016 is \lumireluncert, and it is applied as a normalisation uncertainty for all processes estimated using simulation. It is derived, following a methodology similar to that detailed in Ref. [Aaboud:2016hhf], from a preliminary calibration of the luminosity scale using beam-separation scans performed in August 2015 and May 2016. A variation in the pile-up reweighting of MC events is included to cover the uncertainty in the ratio of the predicted and measured inelastic cross-sections in the fiducial volume defined by where is the mass of the hadronic system [Aaboud:2016mmw].

Uncertainties associated with charged leptons arise from the trigger selection, the object reconstruction, the identification, and the isolation criteria, as well as the lepton momentum scale and resolution. These are estimated by comparing () events in data and simulation [Aad:2016jkr, ATLAS-CONF-2016-024]. Correction factors are applied to the simulation to better model the efficiencies observed in data. The charged-lepton uncertainties have a small impact on the analysis.

Uncertainties associated with jets arise from the jet reconstruction and identification efficiencies related to the JES and jet energy resolution, and on the Jet Vertex Tagger efficiency [PERF-2011-04]. The JES-related uncertainties contain 23 components that are treated as statistically independent and uncorrelatd. The JES and its uncertainty were derived by combining information from test-beam data, LHC collision data (in situ techniques) and simulation [PERF-2016-04]. The many sources of uncertainties related to the in situ calibration using +jets, +jets and multi-jet data were reduced to eight uncorrelated components through an eigen-decomposition. Other components are relativ to jet flavour, pile-up corrections, -dependence and high- jets.

In the reconstruction of quantities used for the BDT, \met is used. The \met calculation depends on the reconstruction of leptons and jets. The uncertainties associated with these objects are therefore propagated to the \met uncertainty estimation. Uncertainties due to soft objects (not included in the calculation of the leptons and jets) are also considered [PERF-2016-07].

Differences between data and simulation in the -tagging efficiency for -jets, -jets and light jets are taken into account using correction factors. For -jets, the corrections are derived from \ttbar events with final states containing two leptons, and the corrections are consistent with unity within uncertainties at the level of a few percent over most of the jet \pt range. The mis-tag rate for -jets is also measured in \ttbar events, identifying hadronic decays of bosons including -jets. For light jets, the mis-tag rate is measured in multi-jet events using jets containing secondary vertices and tracks with impact parameters consistent with a negative lifetime. Systematic uncertainties affecting the correction factors are derived in the \pt and bins used for extracting the correction factors. They are transformed into uncorrelated components using an eigenvector decomposition, taking into account the bin-to-bin correlations [Aad:2015ydr, ATL-PHYS-PUB-2016-012, ATLAS-CONF-2018-001]. For , corrections corresponding to the fixed working point of 70% efficiency are used and a total of 6, 3 and 16 independent uncorrelated eigen-variations are considered as systematic uncertainties for -, - and light jets, respectively. For , corrections for the step-wise efficiencies are used to support the kinematic discriminant and the number of eigen-variations is increased by a factor of five to account for the five -tagging efficiency bins. In addition, uncertainties due to tagging the hadronic decays of -leptons as -jets are considered. For , an additional uncertainty is included due to the extrapolation of scale factors for jets with \GeV, beyond the kinematic reach of the data calibration samples used [ATL-PHYS-PUB-2016-012].

The uncertainty due to different scale choices in the \Hp signal is estimated by varying the renormalisation and factorisation scales up and down by a factor of two. The uncertainty ranges from 7% at low masses to 15% at masses above 1300 \GeV for the \ljets final state, and from 12% to 16.5% for the \dilepton final state. The PDF uncertainty in the modelling is estimated using the PDF4LHC15_30 PDF set [Butterworth:2015oua], which is based on a combination of the CT14 [ct14], MMHT14 [mmht14] and NNPDF3.0 [Ball:2014uwa] PDF sets and contains 30 components obtained using the Hessian reduction method [hessian-reduction-1, hessian-reduction-2, mc-comb-method].

The modelling of the \ttbar+ jets background is one of the largest sources of uncertainty in the analysis and many different components are considered. The uncertainty in the inclusive \ttbar production cross-section at NNLO+NNLL [Czakon:2011xx] is , including effects from varying the factorisation and renormalisation scales, the PDF, the QCD coupling constant , and the top quark mass. Due to the large difference between the 4FS prediction and the various 5FS predictions for the \ttbbb process, an additional 50% normalisation uncertainty is assigned to this background.

The uncertainty due to the choice of NLO generator is derived by comparing the nominal \powheg sample with a sample generated using \SHERPAV2.2.1 with a 5FS PDF. A \powheg sample with the same settings as in the nominal \powheg+\pythia8 sample, but using \herwig[Bahr:2008pv, Bellm:2015jjp] for parton showering, is used to assess the uncertainty due to the choice of parton shower and hadronisation model. Furthermore, the uncertainty due to the modelling of initial- and final-state radiation is evaluated with two different \powheg+\pythia8 samples in which the radiation is increased or decreased by halving or doubling the renormalisation and factorisation scales in addition to simultaneous changes to the \hdampparameter and the A14 tune parameters [ATL-PHYS-PUB-2016-004].

For the \ttbin background, an additional uncertainty is assigned by comparing the predictions from \powheg+\pythia8 and \SHERPAwith 4FS. This takes into account the difference between a 5FS inclusive \ttbar prediction at NLO and a 4FS NLO prediction. For the \ttcin background, an additional uncertainty is derived by comparing a \mg5_\amc sample that is interfaced to \HERWIGpp [Bahr:2008pv] with the nominal event sample. In this \mg5_\amc event sample, a three-flavour scheme is employed and the process is generated at the matrix element level [ATL-PHYS-PUB-2016-011] using the CT10F3 PDF set, while in the nominal sample the charm jets are primarily produced in the parton shower. All of these uncertainties, with the exception of the inclusive and \ttbbb cross-sections, are considered to be uncorrelated amongst the \ttbin, \ttcin, and \ttbar+ light samples. For the modelling of the \ttbin backgrounds, the alternative samples are reweighted to the NLO prediction of from \SHERPAbefore the uncertainty is evaluated.

In addition, uncertainties due to the reweighting to the \SHERPANLO prediction of are considered. For these uncertainties, the \ttbin is reweighted to different \SHERPApredictions with modified scale parameters, in particular where the renormalisation scale is varied up and down by a factor of two, where the functional form of the resummation scale is changed to and where a global scale choice is used. Two alternative PDF sets, MSTW2008NLO [Martin:2009iq] and NNPDF2.3NLO [Ball:2014uwa], are used, and uncertainties in the underlying event and parton shower are estimated from samples with an alternative set of tuned parameters for the underlying event and an alternative shower recoil scheme. Due to the absence of -jets from multi-parton interactions and final-state gluon radiation in the prediction from \SHERPA, a 50% uncertainty is assigned to the \ttb (MPI/FSR) category based on studies of different sets of UE tunes. An uncertainty due to the reweighting of the leading jet \pt is determined by comparing a reweighted event sample with an event sample without reweighting. Because the reweighting changes the normalisation for jet  \GeV by 15%, an additional normalisation uncertainty of 15% is applied in this region. The reweighting factors are derived from the CR with exactly four jets and at least two -tagged jets and applied to higher jet multiplicity bins. However, the effect of this extrapolation is expected to be small and is covered by the above uncertainties.

An uncertainty of is assigned to the total cross-section for single top-quark production [Kidonakis:2010ux, Kidonakis:2010tc, Kidonakis:2011wy], uncorrelated between and -channel production. An additional uncertainty due to initial- and final-state radiation is estimated using samples with factorisation and renormalisation scale variations and appropriate variations of the Perugia 2012 set of tuned parameters. The parton showering and hadronisation modelling uncertainties in the single-top and -channel production are estimated by comparing with samples where the parton shower generator is \HERWIGpp instead of \PYTHIAV6.428. The uncertainty in the interference between and \ttbar production at NLO [Frixione:2008yi] is assessed by comparing the default ‘diagram removal’ scheme with an alternative ‘diagram subtraction’ scheme [Frixione:2008yi, PowhegWt].

The uncertainty arising from \ttV generation is estimated by comparison with samples generated with \SHERPA. The uncertainty in the \ttV production cross-section is about 15%, taken from the NLO predictions [Kardos:2011na, Lazopoulos:2008de, Campbell:2012dh, deFlorian:2016spz], treated as uncorrelated between and with PDF and QCD scale variations.

The \ttH modelling uncertainty is assessed through an uncertainty in the cross-section, uncorrelated between QCD (%) and the PDFs (%) [deFlorian:2016spz, Raitio:1978pt, Beenakker:2002nc, Dawson:2003zu, Yu:2014cka, Frixione:2015zaa], and the modelling of the parton shower and hadronisation by comparing \pythia8 with \HERWIGpp. The minor backgrounds, \tHjb and \WtH are treated as one background and its cross-section uncertainty is 6% due to PDF uncertainties and another 10% due to factorisation and renormalisation scale uncertainties [deFlorian:2016spz].

The uncertainties from the data-driven estimation of non-prompt leptons are based on a comparison between data and the non-prompt lepton estimates in CRs. A 50% uncertainty is assigned in the +jets final state. In the \dilepton final state, where all backgrounds with one or no prompt leptons fall into this category, including +jets and single top production, an uncertainty of is assigned.

An uncertainty of 40% is assumed for the +jets cross-section, uncorrelated between jet bins, with an additional 30% for +HF jets, uncorrelated for two, three and more than three HF jets. These uncertainties are derived from variations of the renormalisation and factorisation scales and matching parameters in \SHERPAsimulations. An uncertainty in +jets of 35% is applied, uncorrelated among jet bins in the \dilepton final state. This uncertainty accounts for both the variation of the scales and matching parameters in \SHERPAsimulations and the data-driven correction factors applied to the +HF jets component. In the \dilepton final state, only the jets component is estimated separately, and the jets background is included in the estimation of the background from non-prompt leptons.

7 Statistical analysis

In order to test for the presence of an \hplus signal, a binned maximum-likelihood fit to the data is performed simultaneously in all categories, and each mass hypothesis is tested separately. The inputs to the fit include the number of events in the CRs and the binned BDT output in the SRs. Two initially unconstrained fit parameters are used to model the normalisation of the \ttbin and \ttcin backgrounds. The procedures used to quantify the level of agreement with the background-only or background-plus-signal hypothesis and to determine exclusion limits are based on the profile likelihood ratio test and the method [Asimov, CLs, Junk:1999kv]. The parameter of interest is the signal strength, , defined as the product of the production cross-section and the branching ratio .

To estimate the signal strength, a likelihood function, , is constructed as the product of Poisson probability terms. One Poisson term is included for every CR and every bin of the BDT distribution in the SRs. The expected number of events in the Poisson terms is a function of , and a set of nuisance parameters, . The nuisance parameters encode effects from the normalisation of backgrounds, including two free normalisation factors for the \ttgeb and \ttgec backgrounds, the systematic uncertainties and one parameter per bin to model statistical uncertainties in the simulated samples. All nuisance parameters are constrained with Gaussian or log-normal terms. There are about 170 nuisance parameters considered in the fit, the number varying slightly across the range of mass hypotheses.

To extract the exclusion limit on , the following test statistic is used:

The values of the signal strength and nuisance parameters that maximise the likelihood function are represented by and , respectively. For a given value of , the values of the nuisance parameters that maximise the likelihood function are represented by .

8 Results

Tables 3 and 4 show the post-fit event yields under the background-plus-signal hypothesis for a signal mass . A value of  pb is obtained from the fit. The corresponding post-fit distributions of the BDT discriminant in the SRs are shown in Figures 6 and 7 for a 200 \GeV \hplus mass hypotheses for the \ljets and \dilepton final state, respectively.

Process CR \FiveJexTwoBex SR \FiveJexThreeBex SR \FiveJexFourBin CR \SixJinTwoBex SR \SixJinThreeBex SR \SixJinFourBin
\ttgeb 15 300  2300 7400  1000 750  110 17100  2800 11 100  1500 2410  260
\ttgec 47 000  12 000 6400  1700 260  80 55 000  11 000 9400  2000 450  180
\ttbar + light 226 000  11 000 12 200  1100 89  35 132 000  10 000 8500  1100 260  120
Non-prompt leptons 15 000  6000 600  500 11  8 13 000  6000 700  400 4  5
\ttW 340  50 29  4 0.66  0.22 540  80 72  11 5.0  1.2
\ttZ 390  50 78  10 12.2  2.2 720  90 183  23 50  7
Single top 8900  2400 690  210 23  13 5400  1800 640  260 53  31
Other top 328  27 28.2  2.6 3.1  0.6 183  20 46  11 14  5
Diboson 410  210 29  15 2.0  2.1 340  170 37  19 4.3  2.5
+ jets 9000  4000 540  240 16  9 5200  2100 470  200 27  12
+ jets 2100  600 104  35 4.9  1.8 1300  400 130  40 11  4
\ttH 252  24 127  13 30  4 520  50 315  32 117  16
\tH 19.5  2.4 10.6  1.3 2.21  0.32 27.2  3.5 15.7  2.0 5.0  0.7
Total 328 000  7000 28 400  900 1220  60 233 000  6000 31 800  800 3410  150
Data 334 813 29 322 1210 234 053 32 151 3459
\hplus (200 \GeV) 470  50 220  23 25.3  3.3 340  50 235  34 60  9
\hplus (800 \GeV) 630  90 390  70 56  12 1230  190 1020  170 350  70
Table 3: Event yields of the SM background processes and data in all categories of the \ljets final state, after the fit to the data under the background-plus-signal hypothesis ( \GeV). The expected event yields for the \hplus signal masses of 200 \GeV and 800 \GeV are shown with pre-fit uncertainties and assuming a cross-section times branching ratio of 1 pb. The quoted uncertainties include both the statistical and systematic components. The uncertainties take into account correlations and constraints of the nuisance parameters. ‘Other top’ includes contributions from as well as - and -channel single top production.
Process CR \ThreeJexTwoBex SR/CR \ThreeJexThreeBex CR \FourJinTwoBex SR \FourJinThreeBex SR \FourJinFourBin
\ttgeb 2330  330 940  130 3300  500 2050  280 322  35
\ttgec 6100  1300 520  140 9900  2000 1310  290 30  14
\ttbar + light 50 700  2300 260  70 32 500  2100 420  120 4  5
Non-prompt leptons 420  110 6.7  2.4 620  160 48  13 2.2  0.8
\ttW 48  7 1.48  0.17 129  7 9.8  1.1 0.55  0.21
\ttZ 43  5 5.8  1.1 174  10 32.9  2.0 7.0  1.3
Single top 1700  500 40  12 1110  330 63  26 3.9  2.0
Other top 3.9  0.5 0.12  0.05 21.8  3.5 5.8  2.2 2.0  0.9
Diboson 36  4 1.2  0.4 46  6 3.1  0.9 0.48  0.28
+ jets 1600  500 42  16 1300  400 82  29 5.3  2.0
\ttH 26.2  1.3 8.5  0.5 116  6 52.2  3.5 16.0  1.9
\tH 1.95  0.27 0.42  0.10 5.7  0.7 2.14  0.32 0.48  0.09
Total 62 800  2800 1810  110 49 300  2300 4060  200 390  28
Data 62 399 1774 48 356 4047 376
\hplus (200 \GeV) 92  12 27  4 72  12 49  8 9.0  1.6
\hplus (800 \GeV) 70  12 32  7 212  33 157  27 44  9
Table 4: Event yields of the SM background processes and data in all categories of the \dilepton final state, after the fit to the data under the background-plus-signal hypothesis ( \GeV). The expected event yields for the \hplus signal masses of 200 \GeV and 800 \GeV are shown with pre-fit uncertainties and assuming a cross-section times branching ratio of 1 pb. The quoted uncertainties include both the statistical and systematic components. The uncertainties take into account correlations and constraints of the nuisance parameters. ‘Other top’ includes contributions from as well as - and -channel single top production.

[] []

[] []

Figure 6: Distributions of the BDT output after the fit to the data in the four SRs of the \ljets final state: (a) \FiveJexThreeBex, (b) \SixJinThreeBex, (c) \FiveJexFourBin and (d) \SixJinFourBin for the 200 \GeV mass hypothesis. Each background process is normalised according to its post-fit cross-section. The includes contributions from \ttW, \ttZ and \ttH. The total prediction of the BDT distributions includes cases where the signal obtained from the fit is negative. For this particular mass point the fitted signal strength is  pb. The pre-fit signal distribution is shown superimposed as a dashed line with arbitrary normalisation. The lower panels display the ratio of the data to the total prediction. The hatched bands show the post-fit uncertainties.

[] []

[]

Figure 7: Distributions of the BDT output after the fit to the data in the three SRs of the \dilepton final state: (a) \ThreeJexThreeBex, (b) \FourJinThreeBex and (c) \FourJinFourBin for the 200 \GeV mass hypothesis. Each background process is normalised according to its post-fit cross-section. The includes contributions from \ttW, \ttZ and \ttH. The total prediction of the BDT distributions includes cases where the signal obtained from the fit is negative. For this particular mass point the fitted signal strength is  pb. The pre-fit signal distribution is shown superimposed as a dashed line with arbitrary normalisation. The lower panels display the ratio of the data to the total prediction. The hatched bands show the post-fit uncertainties.

A summary of the systematic uncertainties is given in Table 5. Depending on the particular \hplus mass hypothesis, the total systematic uncertainty is dominated by the uncertainties in the modelling of the \ttbin background, the jet flavour-tagging uncertainties and the uncertainties due to the limited size of simulated event samples.

Uncertainty Source [pb] [pb]
Jet flavour tagging 0. 70 0. 050
\ttgebmodelling 0. 65 0. 008
Jet energy scale and resolution 0. 44 0. 031
+light modelling 0. 44 0. 019
MC statistics 0. 37 0. 044
\ttgecmodelling 0. 36 0. 032
Other background modelling 0. 36 0. 039
Luminosity 0. 24 0. 010
Jet-vertex assoc., pile-up modelling 0. 10 0. 006
Lepton, \MET, ID, isol., trigger 0. 08 0. 003
\Hp modelling 0. 03 0. 006
Total systematic uncertainty 1. 4 0. 11
\ttgebnormalisation 0. 61 0. 022
\ttgecnormalisation 0. 28 0. 012
Total statistical uncertainty 0. 69 0. 050
Total uncertainty 1. 5 0. 12
Table 5: The summary of the effects of the systematic uncertainties on the signal strength parameter, , for the combination of the \ljets and \dilepton final states is shown for an \hplus signal with a mass of 200 and 800 \GeV. Due to correlations between the different sources of uncertainty, the total systematic uncertainty can be different from the sum in quadrature of the individual sources. The normalisation factors for both \ttgeb and \ttgec are included in the statistical component. The total uncertainty corresponds to a best-fit value of of  pb at  \GeV and  pb at  \GeV. The expected upper limit on is 3.05 pb at  \GeV and 0.26 pb at  \GeV.

The 95% confidence level (CL) upper limits on using the method are presented in Figure 8. The observed (expected) 95% CL upper limits on the production cross-section times the branching ratio range from  pb at to  pb at . The compatibility of the SM hypothesis with the results obtained from the fit to the data is tested. The largest deviation from the SM hypothesis is observed at 300 \GeV. Given that a negative is observed under this mass hypothesis, the test statistic is used to quantify the deviation of the fitted result from the SM expectation. A local value of 1.13% is obtained at 300 \GeV, corresponding to the probability to obtain a deviation at least as large as the one observed in data provided that only SM processes are present.

Figure 8: Expected and observed limits for the production of \hplustb in association with a top quark and a bottom quark. The bands surrounding the expected limit show the 68% and 95% confidence intervals. The limits are based on the combination of the \ljets and \dilepton final states. Theory predictions are shown for three representative values of \tanbeta in the \mhmodm benchmark scenario [Carena:2013ytb]. Uncertainties in the predicted \hplus cross-sections or branching ratios are not considered.

Figure 9 shows 95% CL exclusion limits set on \tanbeta for the \mhmodm scenario of the MSSM [Carena:2013ytb, Heinemeyer:2013tqa, deFlorian:2016spz] and the hMSSM [Djouadi:2013uqa, Djouadi:2015jea, Bagnaschi:2015hka]. Beyond tree level, the Higgs sector is affected by the choice of parameters in addition to Higgs boson masses and \tanbeta. For the \mhmodm benchmark scenario the top-squark mixing parameter is chosen such that the mass of the lightest CP-even Higgs boson, , is close to the measured mass of the Higgs boson that was discovered at the LHC. In the hMSSM scenario, instead of adjusting the parameters of soft supersymmetry breaking, the value of is used to predict the masses and couplings of the MSSM Higgs bosons.

For \hplus masses of 200–920 \GeV (200–965 \GeV), the observed exclusion of low values of \tanbeta at 95% CL is in the range 0.5–1.91 (0.5–1.95) for the \mhmodm (hMSSM) scenario. The most stringent limits on \tanbeta are set for \hplus masses around 250 \GeV. High values of \tanbeta between 36 and 60 are excluded in the \hplus mass range 200–520 \GeV (220–540 \GeV) for the \mhmodm (hMSSM) scenario. The most stringent exclusion, , is at 300 \GeV for both the \mhmodm and hMSSM benchmark scenarios. In the \mhmodm scenario for , the observed (expected) exclusion of \hplus masses is  \GeV ( \GeV).

In comparison with a previous search for production followed by decays [Aad:2015typ], more stringent limits on \hplus masses for particular models and parameter choices can be set. The analysis reach is increased and now also includes \Hplus masses between 600 \GeV and 2 \TeV. The excluded region of parameter space for the model-dependent interpretation is extended significantly for low \tanbeta and an additional excluded region is added at high \tanbeta.

[] []

Figure 9: Expected and observed limits on \tanbeta as a function of in the \mhmodm [Carena:2013ytb] (left) and the hMSSM [Djouadi:2013uqa] (right) scenarios of the MSSM. Limits are shown for \tanbeta values in the range of 0.5–60, where predictions are available from both scenarios. The bands surrounding the expected limits show the 68% and 95% confidence intervals. The limits are based on the combination of the \ljets and \dilepton final states. The production cross-section of \ttH and \tH, as well as the branching ratios of the \Higgs, are fixed to their SM values at each point in the plane. Uncertainties in the predicted \hplus cross-sections or branching ratios are not considered.

The ATLAS Collaboration has also set limits on the production of \hplus using the decay with the same data [ATLAS-HIGG-2016-11]. The final state can be used to set limits at high \tanbeta which are more stringent than those from the final state, and to probe \hplus masses below 200 \GeV, in both the \mhmodm and hMSSM scenarios. Figure 10 shows a superposition of the limits from the two final states, where the limits from the final state exclude a larger portion of the parameter space at high \tanbeta and low \hplus masses than the limits alone.

[] []

Figure 10: Expected and observed limits on \tanbeta as a function of in the \mhmodm [Carena:2013ytb] (left) and the hMSSM [Djouadi:2013uqa] (right) scenarios of the MSSM. Limits are shown for \tanbeta values in the range of 0.5–60, where predictions are available from both scenarios. The limits are a superposition of the results obtained in the analysis presented here, and the ATLAS limits derived from the decay [ATLAS-HIGG-2016-11]. The expected limits from the final state are shown as the horizontally hatched area, with the observed limit as a dash-dotted curve. The expected limits from the final state are shown as diagonally hatched areas, with the observed limit as dashed lines. At low \tanbeta, the strongest limits are from the final state, whereas the exclusions at high \tanbeta and low \hplus masses are obtained from the final state. The exclusion limits for the hMSSM scenario are shown only for  \GeV, where the corresponding theoretical predictions are available.

9 Conclusions

A search for charged Higgs bosons is performed using a data sample corresponding to an integrated luminosity of \lumi from collisions at , recorded by the ATLAS detector at the LHC. The search for is performed in the \hplus mass range 200–2000 \GeV. The analysis uses multivariate techniques in the signal regions to enhance the separation of signal from background and utilises control regions to reduce the effect of large uncertainties in background predictions.

No significant excess above the expected SM background is found and observed (expected) 95% CL upper limits are set on the production cross-section times the branching ratio , which range from = 2.9 (3.0) pb at to = 0.070 (0.077) pb at .

In the context of the \mhmodm (hMSSM) scenario of the MSSM, some values of \tanbeta, in the range 0.5–1.91 (0.5–1.95), are excluded for \hplus masses of 200–920 (200–965) \GeV. For \hplus masses between 200 and 520 \GeV (220 and 540 \GeV), high values of \tanbeta are excluded, e.g.  is excluded at 300 \GeV.

Additionally, taking into consideration the decay, even stricter exclusions can be made at high \tanbeta and low \hplus masses. In the context of the hMSSM, the \hplus mass range up to 1100 \GeV is excluded at , and all \tanbeta values are excluded for \mhp below 160 \GeV.

Acknowledgements

We thank CERN for the very successful operation of the LHC, as well as the support staff from our institutions without whom ATLAS could not be operated efficiently.

We acknowledge the support of ANPCyT, Argentina; YerPhI, Armenia; ARC, Australia; BMWFW and FWF, Austria; ANAS, Azerbaijan; SSTC, Belarus; CNPq and FAPESP, Brazil; NSERC, NRC and CFI, Canada; CERN; CONICYT, Chile; CAS, MOST and NSFC, China; COLCIENCIAS, Colombia; MSMT CR, MPO CR and VSC CR, Czech Republic; DNRF and DNSRC, Denmark; IN2P3-CNRS, CEA-DRF/IRFU, France; SRNSFG, Georgia; BMBF, HGF, and MPG, Germany; GSRT, Greece; RGC, Hong Kong SAR, China; ISF and Benoziyo Center, Israel; INFN, Italy; MEXT and JSPS, Japan; CNRST, Morocco; NWO, Netherlands; RCN, Norway; MNiSW and NCN, Poland; FCT, Portugal; MNE/IFA, Romania; MES of Russia and NRC KI, Russian Federation; JINR; MESTD, Serbia; MSSR, Slovakia; ARRS and MIZŠ, Slovenia; DST/NRF, South Africa; MINECO, Spain; SRC and Wallenberg Foundation, Sweden; SERI, SNSF and Cantons of Bern and Geneva, Switzerland; MOST, Taiwan; TAEK, Turkey; STFC, United Kingdom; DOE and NSF, United States of America. In addition, individual groups and members have received support from BCKDF, CANARIE, CRC and Compute Canada, Canada; COST, ERC, ERDF, Horizon 2020, and Marie Skłodowska-Curie Actions, European Union; Investissements d’ Avenir Labex and Idex, ANR, France; DFG and AvH Foundation, Germany; Herakleitos, Thales and Aristeia programmes co-financed by EU-ESF and the Greek NSRF, Greece; BSF-NSF and GIF, Israel; CERCA Programme Generalitat de Catalunya, Spain; The Royal Society and Leverhulme Trust, United Kingdom.

The crucial computing support from all WLCG partners is acknowledged gratefully, in particular from CERN, the ATLAS Tier-1 facilities at TRIUMF (Canada), NDGF (Denmark, Norway, Sweden), CC-IN2P3 (France), KIT/GridKA (Germany), INFN-CNAF (Italy), NL-T1 (Netherlands), PIC (Spain), ASGC (Taiwan), RAL (UK) and BNL (USA), the Tier-2 facilities worldwide and large non-WLCG resource providers. Major contributors of computing resources are listed in Ref. [ATL-GEN-PUB-2016-002].

Appendix

Appendix A BDT input variables

In this appendix, the full list of variables used as inputs to the BDTs, described in Section 5, is reported.

\ljets channel
Leading jet transverse momentum
Invariant mass of pair of -tagged jets with smallest
Transverse momentum of fifth jet
Second Fox–Wolfram moment [Bernaciak2013] calculated using all jets and leptons
Average between all -tagged jet pairs in the event
between the lepton and the -tagged jet pair with smallest
Invariant mass of the non--tagged jet-pair with minimum
\Htjets Scalar sum of all jets transverse momenta
Invariant mass of the -tagged jet pair with maximum transverse momentum
Largest invariant mass of any two -tagged jets
Largest invariant mass of any three jets
Kinematic discriminant based on mass templates (for )
\dilepton channel, \ThreeJexThreeBex \FourJinThreeBex \FourJinFourBin
Inv. mass of the jet and -tagged jet with largest \pt
Energy difference between the third jet and the subleading lepton
Energy of third jet
Inv. mass difference between and
Angular difference between subleading jet and
\pt of leading -tagged jet
\pt of the pair of lepton and -tagged jet with largest
Inv. mass of the pair of lepton and -tagged jet with smallest
Energy difference between the leading -tagged jet and
Inv. mass difference between and
Inv. mass difference between and
\pt difference between leading and third jet
Smallest invariant mass of any -tagged jet pair
Smallest invariant mass of any pair of lepton and -tagged jet
\pt of
Angular difference between and
Scalar sum of all jets and leptons transverse energy
\dilepton channel, \ThreeJexThreeBex \FourJinThreeBex \FourJinFourBin
\pt of the pair of lepton and -tagged jet with smallest
\pt difference between leading and third jets
Inv. mass difference between and
\pt of the pair of lepton and -tagged jet with smallest
Inv. mass of the jet pair with smallest
\pt difference between leading jet and
\pt of
Energy difference between and
Energy of the leading jet
Maximum \pt of any jet pair
Inv. mass of
\pt of the lepton--jet pair with smallest separation in
\pt difference between subleading lepton and
\pt difference between subleading lepton and
\pt difference between subleading lepton and
\pt difference between leading jet and
Energy difference between leading lepton and
Smallest invariant mass of any -tagged jet pair
\Htall Scalar sum of all jets and leptons transverse momenta
\pt of
\pt difference between subleading -tagged jet and
\pt difference between subleading jet and
Energy difference between third jet and
Inv. mass difference between and

Table 6: Input variables to the classification BDT in the \ljets and \dilepton channels. The symbols , , , and \met represent the four-momenta of jets, -tagged jets, non--tagged jets, the lepton and the missing transverse momentum. All numbered indices refer to ordering in transverse momentum, with 1 as leading. The SRs where the variables are used are indicated for the \dilepton channel. In the \ljets channels the variables are used for all channels. In the \ljets channel the discriminant, , is only used for charged Higgs boson masses . For the \dilepton channel a very large set of kinematic variables using combinations of the analysis objects was examined, and approximately ten optimal variables were selected for each SR independently for the low-mass region ( \GeV) and the high-mass region ( \GeV).
\printbibliography

The ATLAS Collaboration

M. Aaboud, G. Aad, B. Abbott, O. Abdinov, B. Abeloos, D.K. Abhayasinghe, S.H. Abidi, O.S. AbouZeid, N.L. Abraham, H. Abramowicz, H. Abreu, Y. Abulaiti, B.S. Acharya, S. Adachi, L. Adamczyk, J. Adelman, M. Adersberger, A. Adiguzel, T. Adye, A.A. Affolder, Y. Afik, C. Agheorghiesei, J.A. Aguilar-Saavedra, F. Ahmadov, G. Aielli, S. Akatsuka, T.P.A. Åkesson, E. Akilli, A.V. Akimov, G.L. Alberghi, J. Albert, P. Albicocco, M.J. Alconada Verzini, S. Alderweireldt, M. Aleksa, I.N. Aleksandrov, C. Alexa, T. Alexopoulos, M. Alhroob, B. Ali, G. Alimonti, J. Alison, S.P. Alkire, C. Allaire, B.M.M. Allbrooke, B.W. Allen, P.P. Allport, A. Aloisio, A. Alonso, F. Alonso, C. Alpigiani, A.A. Alshehri, M.I. Alstaty, B. Alvarez Gonzalez, D. Álvarez Piqueras, M.G. Alviggi, B.T. Amadio, Y. Amaral Coutinho, L. Ambroz, C. Amelung, D. Amidei, S.P. Amor Dos Santos, S. Amoroso, C.S. Amrouche, C. Anastopoulos, L.S. Ancu, N. Andari, T. Andeen, C.F. Anders, J.K. Anders, K.J. Anderson, A. Andreazza, V. Andrei, C.R. Anelli, S. Angelidakis, I. Angelozzi, A. Angerami, A.V. Anisenkov, A. Annovi, C. Antel, M.T. Anthony, M. Antonelli, D.J.A. Antrim, F. Anulli, M. Aoki, J.A. Aparisi Pozo, L. Aperio Bella, G. Arabidze, J.P. Araque, V. Araujo Ferraz, R. Araujo Pereira, A.T.H. Arce, R.E. Ardell, F.A. Arduh, J-F. Arguin, S. Argyropoulos, A.J. Armbruster, L.J. Armitage, A Armstrong, O. Arnaez, H. Arnold, M. Arratia, O. Arslan, A. Artamonov, G. Artoni, S. Artz, S. Asai, N. Asbah, A. Ashkenazi, E.M. Asimakopoulou, L. Asquith, K. Assamagan, R. Astalos, R.J. Atkin, M. Atkinson, N.B. Atlay, K. Augsten, G. Avolio, R. Avramidou, M.K. Ayoub, G. Azuelos, A.E. Baas, M.J. Baca, H. Bachacou, K. Bachas, M. Backes, P. Bagnaia, M. Bahmani, H. Bahrasemani, A.J. Bailey, J.T. Baines, M. Bajic, C. Bakalis, O.K. Baker, P.J. Bakker, D. Bakshi Gupta, E.M. Baldin, P. Balek, F. Balli, W.K. Balunas, J. Balz, E. Banas, A. Bandyopadhyay, S. Banerjee, A.A.E. Bannoura, L. Barak, W.M. Barbe, E.L. Barberio, D. Barberis, M. Barbero, T. Barillari, M-S. Barisits, J. Barkeloo, T. Barklow, N. Barlow, R. Barnea, S.L. Barnes, B.M. Barnett, R.M. Barnett, Z. Barnovska-Blenessy, A. Baroncelli, G. Barone, A.J. Barr, L. Barranco Navarro, F. Barreiro, J. Barreiro Guimarães da Costa, R. Bartoldus, A.E. Barton, P. Bartos, A. Basalaev, A. Bassalat, R.L. Bates, S.J. Batista, S. Batlamous, J.R. Batley, M. Battaglia, M. Bauce