您当前的位置: 首页 > 网页快照
Vol. 38 (2007) ACTA PHYSICA POLONICA B No 2OBSERVABILITY OF SAME-CHARGE LEPTONTOPOLOGIES IN FULLY LEPTONIC TOP QUARKPAIR EVENTS IN CMS
Vol. 38 (2007) ACTA PHYSICA POLONICA B No 2OBSERVABILITY OF SAME-CHARGE LEPTONTOPOLOGIES IN FULLY LEPTONIC TOP QUARKPAIR EVENTS IN CMS ?Steven LowetteVrije Universiteit Brussel — IIHEPleinlaan 2, 1050 Brussel, Belgium(Received November 15, 2006)At the Large Hadron Collider dileptonic tt?(+jets) events can be selectedwith a relatively high signal-to-noise ratio and efficiency, with backgroundevents produced via Standard Model diagrams. Within the clean sampleof these events, both isolated leptons have an opposite electric charge. Inseveral models beyond the Standard Model tt/t?t?(+jets) topologies are pre-dicted, kinematically similar to the Standard Model tt?(+jets) signature,where both leptons have an equal electric charge. Such a signal of newphysics can be diluted by the mis-identification of the leptons or theirelectric charge in Standard Model tt?(+jets) events. The observability ofan excess of same-charge dilepton signals above the mis-reconstruction ofthe Standard Model background is presented, assuming the same topology.With an integrated luminosity of 30 fb?1, a same-charge dilepton signatureof pp→ tt/t?t? events with a cross section larger than 1.2 pb is visible in themeasurement of the ratio between same-charge and opposite-charge leptonpair events [J. D’Hondt, S. Lowette, G. Hammad, J. Heyninck, P. Van Mul-ders, “Observability of same-charge lepton topology in dileptonic events tt?”,CERN-CMS-NOTE-2006-065.]PACS numbers: 12.90.+b1. IntroductionIn the Compact Muon Solenoid (CMS) experiment [2] at the LargeHadron Collider (LHC) millions of tt? events will be produced. The branch-ing fraction of the top quark is dominated by the t→ Wb decay. When bothW bosons decay leptonically, the so-called dilepton channel, a clear signalof two leptons and two b-quarks is produced. This topology can be distin-guished with a high purity from other Standard Model processes, yieldinga clean sample to search for deviations from Standard Model predictions.? Presented at the “Physics at LHC” Conference, Kraków, Poland, July 3–8, 2006.(469)470 S. LowetteThe Standard Model (SM) predicts that only tt? events will be producedin proton collisions and no tt or t?t? should be observed. Therefore the twoleptons in the dilepton final state should have an opposite electric charge.Lepton identification and the reconstruction of the electric charge of a trackis not fully efficient, however, and hence the SM tt? events can induce a fakesignal of same-sign tt or t?t? events. In this article the Standard Model esti-mation is performed of the ratio of the amount of events with leptons withsame-sign versus opposite-sign electric charge. From this ratio the minimalcross-section for the inclusive pp → tt/t?t? process is determined, needed fora 5σ excess above the Standard Model expectation, as a function of theintegrated luminosity collected by the CMS experiment [1].2. Same-charge top quarks in the Standard Model and beyondThe LHC will be a “top factory”, with tt? production reaching a crosssection of 830 pb. At the low LHC luminosity expected in the first yearsof operation, corresponding to an integrated luminosity of 10 fb?1/y, over8 million tt? pairs will be produced each year. Contrary to the situation at theTevatron collider, top events at the LHC suffer from lower SM backgrounds,because σLHC(tt?)/σTeV(tt?) ? 100 while σLHC(W/Z)/σTeV(W/Z) ? 10.The decay of the top quark t→Wb has a branching fraction of ? 100%in the SM. The signature of a tt? pair is hence given by the decay topologyof the W pair in addition to the two b-quarks in the final state. This studyfocuses on the dilepton final states. Each of the ee and ?? final states take1.2% of the total tt? cross section, the e? final state twice as much.The Standard Model predicts that no tt or t?t? should be observed. Stilla same-charge final state can be faked due to wrong charge determinationand muon and electron identification inefficiencies. Certain models beyondthe Standard Model, on the other hand, predict direct production of tt or t?t?,e.g. from gluino pairs in supersymmetry with light stop and g?g? → t?t?X → ttXdecays [3], from FCNC with Z in the SM or Z ′ or top-Higgs in technicolourmodels [4], or from technipion production tπ±t → ttb? or tπ0t → ttc? [5].3. Event reconstruction and selectionThe key component of this analysis concerns the reconstruction and iden-tification of the isolated leptons in the final state. Standard CMS o?ineelectron and muon reconstruction is followed by the identification methoddescribed in [6]. A channel-dependent likelihood ratio L is determined foreach lepton. This likelihood ratio is conceived to efficiently identify thecorrect lepton from the leptonic t→Wb→ ?νb decay.Observability of Same-Charge Lepton Topologies . . . 471The following observables are combined:? the lepton’s transverse momentum;? the isolation energy, calculated with calorimeter deposits near the lep-ton, at the azimuthal side where neutrals in jets are expected;? the isolation pT, being the sum of transverse track momenta aroundthe lepton;? the isolation angle, defined as the angle to the closest jet;? the association significance to the primary vertex;? a reconstruction quality variable for electrons.This method greatly suppresses leptons arising from fake leptons andfrom real leptons in heavy flavour jets. The combined likelihood ratio dis-tributions are shown in Fig. 1 for electrons and muons respectively. Threeclasses of leptons are distinguished: correctly identified leptons (accordingto a matching with the generated truth), with either correct or wrong chargedetermination, and mis-identified leptons.1 1Matched electron Matched muonNon-matched electron Non-matched muonWrong charge electron Wrong charge muon10-1 10-110 2 10-2-10-3 10-310-4 -40 0.1 0.2 0.3 0.4 0.5 0.6 0.7 10 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7Electron Likelihood Muon LikelihoodFig. 1. Combined likelihood ratio distribution for muons (left) and electrons (right).Apart form the leptons, the event selection consists of simple sequen-tial cuts. First the CMS single or double electron or muon trigger cri-teria are applied. Next, 2 jets with ET > 25 GeV are demanded, looselyb-tagged to suppress Z+jets,WW+jets, etc. Finally, 2 leptons are requested(??, ee or e?) with pT > 25 GeV/c and likelihood ratio L > 0.05. The resultsfor this selection are shown in Table I in the three considered final states,472 S. Lowettefor the main SM processes. Contributions from WZ and ZZ productionwere checked to be negligible. All analyzed samples were generated withPYTHIA, and processed with the detailed GEANT4-based CMS detectorsimulation and reconstruction.TABLE IOverview of the selection applied on the considered SM background processes. Theexpected number of events are rescaled to an integrated luminosity of 1 fb?1.tt?→ tt?→ tt?→ tt?→ WW Z + jets?? ?e/ee τ +X otherBefore selection 6915 20745 34606 485973 189952 578033Trigger 6115 16315 17416 100137 41288 2663672 jets ET > 25GeV 4398 11983 13561 93858 20594 66147b-Tag criteria 990 2485 2290 8785 134 2402 leptons identified 888 30 376 803 1.7 732 leptons selected 481.5 0.07 48.4 3.01 0.4 53.3Efficiency (in %) 6.96 0.0003 0.14 0.0006 0.00022 0.0092Opposite charge 481.3 0 48.3 2.19 0 53.3Same charge 0.2 0.07 0.1 0.82 0.4 0tt?→ tt?→ tt?→ tt?→ WW Z + jetsee ??/?e τ +X otherBefore selection 6915 20745 34606 485973 189952 578033Trigger 5355 17075 17416 100137 41288 2663672 jets ET > 25GeV 3961 12420 13561 93858 20594 66147b-Tag criteria 803 2672 2290 8785 134 2402 leptons identified 725 35 454 2284 73 1272 leptons selected 285.0 0.3 37.5 5.2 0.8 53.3Efficiency (in %) 4.12 0.0013 0.11 0.0011 0.00044 0.0092Opposite charge 279.6 0.3 36.8 4.1 0.4 46.7Same charge 5.4 0 0.7 1.1 0.4 6.7tt?→ tt?→ tt?→ tt?→ WW Z + jetse? ee/?? τ +X otherBefore selection 13830 13830 34606 485973 189952 578033Trigger 10960 11470 17416 100137 41288 2663672 jets ET > 25GeV 8022 8359 13561 93858 20594 66147b-Tag criteria 1683 1793 2290 8785 134 2402 leptons identified 1501 66 822 3002 30.2 202 leptons selected 722.7 0.9 85.2 6.3 0.4 0Efficiency (in %) 5.23 0.0065 0.25 0.0013 0.00022 0Opposite charge 715.5 0.9 83.8 4.9 0 0Same charge 7.2 0 1.3 1.4 0.4 0Observability of Same-Charge Lepton Topologies . . . 4734. ResultsFrom the number of selected events, the ratioN++,??R =N+?is determined, with N++,?? and N+? the amount of remaining events withsame-charge and opposite-charge leptons, respectively. For 10 fb?1 the Stan-dard Model expectation isR?? = 0.0027 ± 0.0007 ,Ree = 0.0389 ± 0.0033 ,Re? = 0.0128 ± 0.0013 .For each decay channel, using the uncertainty on the ratio R, the inclu-sive cross-section of the process pp → tt/t?t? can be determined, needed toobtain a 5σ excess above the SM expectation. It is assumed that a signal be-yond the Standard Model has a similar kinematic topology compared to theSM tt? process. In Fig. 2 the significance of the excess in standard deviationsabove the Standard Model prediction of the ratio R is shown as a functionof the inclusive pp→ tt/t?t? cross-section. It has been shown that the smallersignificance with electrons is due to inefficiencies in electron identificationthat can be further improved.Cross-section significance for 30fb-1101?± ?± channele± e± channel?± e± channel10-1 1 2 3 4 5 6 7 8σ(pp → tt / t t ) (pb)Fig. 2. Significance of a same-charge dilepton excess above the SM expectation, asa function of the inclusive pp→ tt/t?t? cross-section.5. Systematic uncertaintiesThe use of the ratio R cancels most of the expected experimental andtheoretical systematics. Several possible remaining sources of systematicswere investigated. The knowledge of the background cross section was shownSignificance474 S. Lowetteto be negligible. A 100% uncertainty on the Z+jets cross section only affectsslightly the ee final state. Uncertainties on R due to the WW cross sectionare also expected negligible. Another possible source of systematics stemsfrom the tt? → τ + X selection efficiency. A 20% variation with respect tothe dilepton final state showed no effect, however.The systematic effect due to uncertainties on the charge determinationis possibly dangerous, and requires an accurate measurement of the chargeidentification efficiency from data. This can be envisaged using the ?10 MZ → ?+?? events per 10 fb?1 after trigger and acceptance cuts. The Z massdistributions can be reconstructed using different (ND events) and same(NS events) charge leptons. For the background an equal amount of eventsNB ? NB = NB can be assumed around the Z mass. The efficiency of theD Scharge identification is then given by ? = NS+2ND . In Fig. 3 the expected2NS+2NDrelative uncertainty on (1 ? ?) is shown for this proposed measurement,as a function of ? itself, for 10 fb?1. With a 10% background contribution,anuncertainty of 4% is expected on a charge mis-identification efficiency of0.1%. Such an uncertainty on the charge determination has negligible effecton the measurement of the ratio R.102101 40%20%10%2%0% Bck10-10.99 0.992 0.994 0.996 0.998 1∈Fig. 3. Relative uncertainty on the charge mis-identification efficiency (1? ?) asa function of ?, for 10 fb?1.6. ConclusionsIt is shown that with a measurement of the ratio R = N++,??/N+? indileptonic top quark pair events, new processes pp → tt/t?t?, kinematicallysimilar to the Standard Model process pp → tt?, can be observed in CMSwith 30 fb?1, if they have a cross section above 1.2 pb. The dimuon channelexhibits the largest sensitivity of the considered decay channels. Most of theexperimental and theoretical systematic uncertainties cancel in the ratio.Potential remaining systematic uncertainties were checked to be negligible.Relative uncertainty on (1-∈) (%)Observability of Same-Charge Lepton Topologies . . . 475REFERENCES[1] J. D’Hondt, S. Lowette, G. Hammad, J. Heyninck, P. Van Mulders, Observ-ability of Same-Charge Lepton Topology in Dileptonic t Anti-t Events, CERN-CMS-NOTE-2006-065.[2] Homepage of the CMS collaboration: http://cms.cern.ch/[3] S. Kraml, A.R. Raklev, Phys. Rev. D73, 075002 (2006) [hep-ph/0512284].[4] F. Larios, F. Penunuri, J. Phys. G30, 895 (2004) [hep-ph/0311056].[5] C.X. Yue, Z.J. Zong, L.L. Xu, J.X. Chen, Phys. Rev., D73, 015006 (2006)[hep-ph/0601058].[6] J. D’Hondt, S. Lowette, J. Heyninck, Electron and Muon Reconstruction inSingle Leptonic t Anti-t Events, CERN-CMS-NOTE-2006-024.
From:
系统抽取对象
机构     
(1)
(10)
(1)
(1)
(7)
(1)
(1)
地理     
(1)
(2)
(1)
(1)
(1)
(1)
(2)
活动
出版物     
(1)
人物     
(1)
(1)
(1)
(1)
系统抽取主题     
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)  
(1)