Search for charged Higgs bosons in ATLAS DanielPelikan,onbehalfoftheATLASCollaborationa UppsalaUniversity,Sweden Abstract. ChargedHiggsbosonsarepredictedinseveralextensionsoftheStandardModel,wheretheHiggs sectorcontainsmorethanonedoubletofcomplexscalars,forinstanceintheMinimalSupersymmetricStandard Model(MSSM).FormH+ < mtop,thedominantproductionmodeforchargedHiggsbosonsattheLHCisvia thedecayt→bH+ofoneofthetopquarksintt¯events.Wepresentresultsonthesearchforsuchlightcharged 2 1 HiggsbosonsintheATLASexperiment,withemphasisonthedecayH+→τν. 0 2 1 Introduction arett¯andsingletop-quarkproduction,W+jetsandmulti- n a jetevents.Allthesebackgroundsaredeterminedwithdata- J OnemajortaskfortheLargeHadronCollider(LHC)and drivenmethods. 3 theATLASexperimentistheexplorationoftheHiggssec- 2 tor in the Standard Model (SM) and beyond it. Charged Higgsbosons,H+ andH−,arepredictedbyseveralexten- g b ] sionsoftheSM,suchasmodelscontainingHiggstriplets x t H+ andTwo-Higgs-DoubletModels(2HDM)[1].Electroweak g e - symmetry breaking through two complex Higgs doublets p leadtofivephysicalstates,outofwhichtwoarecharged. e t¯ W+ TheobservationofchargedHiggsbosonswouldclearlyin- h g [ dicatephysicsbeyondtheSM.Herethetype-II2HDMis ¯b considered,whichisalsotheHiggssectoroftheMinimal Fig.1.FeynmandiagramofthechargedHiggsbosonproduction 1 SupersymmetricStandardModel(MSSM)[2].Forvalues intt¯events. v oftanβ(theratioofthevacuumexpectationvaluesofthe 0 twoHiggsdoublets)largerthan3,typically,chargedHiggs 1 bosonsdecaymainlyviaH+ → τν[3].Inthefollowing,a 7 4 branchingratioB(H+ →τν)=1isconsidered. 2.1 Electron-to-τmisidentificationprobabilitywitha . tag-and-probemethod 1 ThesearchesforthechargedHiggsbosonintt¯events(see 0 Fig.1)withoneortwolightchargedleptons(i.e.electrons Backgroundeventswhereanelectronismisidentifiedasa 2 ormuons)inthefinalstate[4]andinthecasewhereboth hadronicallydecayingτareestimatedfromacontrolsam- 1 : theτandtheWbosondecayhadronically(τ+jetschannel) pleusingadata-driventag-and-probemethod.Inorderto v [5]aredescribed.1.03fb−1ofdatafromproton-protoncol- determinethemisidentificationprobability,Z/γ∗ eventsin Xi lisions,atacentreofmassenergyof √s=7TeV,collected collision data are used. The result in data is compared to withtheATLASexperimentin2011,areanalysed.Upper MonteCarlo(MC)andtheratioisusedasascalefactorin r a limits for B(t → bH±) and the charged Higgs boson pro- MC. duction in the mH+ − tanβ plane (mmHax scenario [6]) are derived. 2.2 Jet-to-τmisidentificationprobabilityfrom photon+jetsevents 2 The τ+jets Final State Inordertodeterminetheprobabilityforajettobemisiden- tifiedasahadronicallydecayingτ,aγ+jetcontrolsample ThisstudydescribesthesearchforthechargedHiggsbo- isused.Likejetsfromthehardprocessesinthedominant sonintt¯eventswhichdecayinto[H+b][W−b¯]andtheninto tt¯background,jetsinthecontrolsampleoriginatemainly [(τ+ +ν)b][qq¯(cid:48)b¯]. For such τ+jets events, the transverse fromquarks.Ameasurementoftheprobabilityforajetto had massmτ isdefinedas: bemisidentifiedasahadronicallydecayingτisperformed T indataandisusedtopredicttheyieldofjet-to-τmisidenti- (cid:113) ficationeventsfromthemostimportantSMbackgrounds. mτ = 2pτEmiss(1−cos∆φ), (1) T T T where∆φistheazimuthalanglebetweentheτ-jetandthe 2.3 Multi-jetbackgroundestimate missingenergydirection.Themainsourcesofbackground TheexpectationsinMCformulti-jeteventshavelargeun- a e-mail:[email protected] certainties, for that reason this background is estimated EPJWebofConferences GeV 25 ATLASPreliminary Data 2011 tan β 60 mmhaxExpected Limit ATLASPreliminary nts / 20 20 TeJe→rut→eτ ττm misiidsid 4500 OEEOthxxbbeppssoeeeer.ccrr vvutteeeenddddc , e±± L±r i12t ma1σσi inσtties e Ev 15 Multi−jets 30 H+(130), B=0.1 H++background 20 10 ∫ L dt=1.03 fb 1 10 Data 2011 ∫ L dt= 1.03 fb 1 0 90 100 110 120 130 140 150 160 5 mH+ [GeV] 0 Fig.4.Limitfromτ+jetseventsforH+productioninthemH+ − 0 50 100 150 200 250 300 tanβplane(mmaxscenario). H m [GeV] T Fig.2.mTdistributionaftereventselection. hadronically(single-lepton)orleptonically(di-lepton).To searchforchargedHiggsbosonsinthesingle-anddi-lepton channels, discriminating variables are used, that allow a ν)0.22 ATLASPreliminary +Hτ→ 00.1.28 Data 2011 ∫ L dt= 1.03 fb 1 doifsWtincotriocnhabregtewdeHeniglgesptboonssofnrso)morτle→ptolνnlsντco(em.gin.ginddireeccatylys × B(0.16 Expected Limit from the W boson decay. The invariant mass m of a b- b) 0.14 Expected ± 1σ bl +H0.12 Expected ± 2σ quarkandalightchargedleptonl(electronormuon)can t→ n B(0.00.81 OD0b sOebrvseedrv Leidmit base:used,ormoreconveniently,thevariablecosθl∗ defined L o0.06 95% C00..0024 cosθ∗ = 2m2bl −1(cid:39) 4pb·pl −1, (2) 0 90 100 110 120 130 140 150 160 l m2 −m2 m2 −m2 mH+ [GeV] top W top W Fig. 3. Expected and observed limit with 95% C.L. of where pb and pl arethefour-momentaoftheb-quarkand B(t → H+b) from τ+jets events as a function of m+H and the thelepton,mtop andmW arethemassofthetop-quarkand comparisonwiththeD0experiment. theW boson. Otherdiscriminatingvariablesaretwonewtransversemass observables[7].Insingle-leptonevents,thechargedHiggs fromdata.ThisisachievedbyfittingitsETmissshapetodata. bosontransversemassmH iscomputedasfollows: Fortheshapestudyacontrolregionisdefinedwheretheτ T identificationandtheb-taggingareinverted. (mH)2 =(cid:18)(cid:113)m2 +(pl +pb +pmiss)2−pb(cid:19)2−(cid:16)pl +pmiss(cid:17)2 T top T T T T T T (3) 2.4 Embeddingmethod In the case of di-lepton events, the generalised charged Theembeddingmethodisusedtoestimatethebackground HiggsbosontransversemassmTH2isobtainedasfollows: fromtrueτjets.Controlsamplesofsingle-top,W+jets,and tt¯events with a muon indata are collected and the detec- mTH2 = max [mTH(pTH+)], (4) {constraints} torsignatureofthismuonisreplacedwiththatofasimu- lated hadronically decaying τ. Then, the reconstruction is where some kinematical constraints (see eq. (9) in [7]) re-appliedonthenewhybridevent. mustbefulfilledandwiththedefinition: mH(p H+))2 =(cid:18)(cid:113)m2 +(p H+ +pb)2−pb(cid:19)2−(cid:16)p H+(cid:17)2. 2.5 Results T T top T T T T (5) InFig.2themT distributionwithallbackgroundsafterall The multi-jet background is estimated using data-driven cuts is shown. Assuming B(H+ → τν) = 1, upper lim- methods,theotherbackgroundsaretakenfromsimulation. its 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 3.1 Single-LeptonEvents upper limits with a 95% Confidence Level (C.L.) on the branchingratio(Fig.3)andtheH+productioninthemmax Inordertoselectsingle-leptontt¯events,thefollowingse- H scenario(Fig.4).Alimitonthebranchingratiobelow10% lectioncutsareapplied: isreached. – exactlyonetrigger-matchedleptonwithE > 25GeV T (electron)or p >20GeV(muon), T 3 Single- and Di-Lepton Final States – at least four jets with pT > 20 GeV and |η| < 2.5, includingexactlytwob-jets, Inthischannel,theτfromthechargedHiggsbosondecays – Emiss >40GeV if|φ |≥π/6, T l,miss leptonically, the W boson on the other side decays either Emiss×|sin(φ )|>20 if|φ |<π/6. T l,miss l,miss HadronColliderPhysicssymposium(HCP-2011) 112200 0.4 vents / 10 GeVvents / 10 GeV1188000000ATLAttSttS i((nwWgPil+rteheW ltiHo m)p+)inamBrrHy(+t ∫ =→L 1d b3t H0= + G)1 =e.0 V130 f%b 1 +→ bHt bound on Br 0000..23..2355 ATLAS±±OEP xb12rpσesσeelcirmvteeiddn aCrLys D∫Ladtat =2 011.10 3s f b= 1 7 TeV EE Z + jets er 0.15 p 46460000 WDQiC b+oD sjeotns 95% C.L. up 0.00.051 Data 2011 90 100 110 120 130 140 150 160 2200 mH+ [GeV] Fig. 7. Expected and observed combined upper limits of 002200 4400 6600 8800 110000 112200114400116600118800 B(t→ H+b)with95%C.L.asafunctionofmH+. mmHH [[GGeeVV]] TT β Fig.5.mH distributionwithcosθ∗<−0.6,mW <60GeV. tan 60 ATLASPreliminary T l T 50 40 225500 VV ∫ GeGe ATLASPreliminary Ldt = 1.03 fb 1 30 mmax ents / 10 ents / 10 220000 ttStt i((nwWgil+tehW tHo )p+) mBrH(+t =→ 1 b3H0 +G) =e V10% 1200 Data 2011 ∫Ldt = 1.03 fb 1 OEOtEEhxxxbbepppssoeeeeer.cccrr vvutttheeeeendddddc , e±±L L±r ii12tm ma1σσiiσinttties EvEv 115500 Z + jets 0 90 100 110 120 130 140 Diboson mH+ [GeV] QCD & W 110000 Data 2011 Fig.8. CombinedlimitsforH+productioninthemH+-tanβplane (mmaxscenario). H 5500 3.3 LimitsontheBranchingRatiooft→bH+ 00 4400 6600 8800 110000112200 114400116600118800220000 mmHTHT22 [[GGeeVV]] The combined limits, from single- and di-lepton events, withaC.L.of95%forthebranchingratioB≡B(t→ bH+) Fig.6.mH distributionforeventswithcosθ∗<−0.6. andthelimitontheproductionforthechargedHiggsbo- T2 l sonareshowninFig.7and8. Afterhavingassignedtwolight-quarkjetsandoneb-jetto thehadronicallydecayingtop-quarkcandidate,theotherb- 4 Conclusion jetisassociatedtotheleptonandthechargedHiggsboson transverse mass is reconstructed, see Fig. 5. To enhance ResultsforthesearchofchargedHiggsbosonsinATLAS the sample with leptonically decaying τs, additional cuts arepresentedusing1.03fb−1of ppcollisiondataat7TeV. on cosθ∗ < −0.6 and mW < 60 GeV are applied. mW is UpperlimitsforthebranchingratioB(t → bH+)between l T T definedasinEq.(1),whereτisreplacedbythelepton. 3%and10%forchargedHiggsbosonmassesintherange of90GeV< m+ <160GeVhavebeenderived,withthe H analysisofmoreATLASdataimprovementsareexpected. 3.2 Di-LeptonEvents References Toselectdi-leptontt¯eventsthefollowingcutsareapplied: – exactly two oppositely charged leptons, including at 1. J.F.GunionandH.E.Haber,Nucl.Phys.B272(1986) least one matched to the trigger with E > 25 GeV 1. T (electron)or p >20GeV(muon), 2. K. Inoue, A. Kakuto, H. Komatsu, and S. Takeshita, T – at least two jets with p > 20 GeV and |η| < 2.5, in- Prog.Theor.Phys.68(1982)927. T cludingexactlytwob-taggedjets, 3. LHC Higgs Cross Section Working Group Collabora- – for ee and µµ events, the invariant mass m must ex- tion,S.Dittmaieretal.,arXiv:hep-ph/1101.0593. ll ceed 15 GeV and must satisfy |m − m | > 10 GeV, 4. ATLAS Collaboration, ATLAS-CONF-2011-151, ll Z togetherwithEmiss >40GeV, https://cdsweb.cern.ch/record/1398187. T – foreµevents,thescalarsumofthetransverseenergies 5. ATLAS Collaboration, ATLAS-CONF-2011-138, oftheselectedleptonsandjetsis(cid:80)E >130GeV. https://cdsweb.cern.ch/record/1383841. T 6. M.S. Carena, S. Heinemeyer, C.E.M. Wagner and G. Afterassignmentofthetwoleptonsandthetwob-jets Weiglein, Eur. Phys. J. C 26 (2003) 601, arXiv:hep- totheircorrectparents,basedoncosθ∗andthedistancebe- ph/0202167. l tweeneachleptonandb-jet,thegeneralisedchargedHiggs 7. E.GrossandO.Vitells,Phys.Rev.D81(2010)055010. bosonmassisreconstructed,seeFig.6.