WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 1 Higgs Physics at ATLAS and CMS Iacopo Vivarelli INFN Pisa and University of Athens On behalf of the ATLAS and CMS collaborations

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 2 Summary Standard Model Higgs Boson: - discovery channels at LHC over the Higgs mass range. Discussion on the main channels in each Higgs mass region. Overall view of the statistical significance for the discovery as a function of M H - SM Higgs parameters measurement: determining the Higgs sector after the discovery MSSM Higgs: discovery perspectives for the LHC experiments All the results assume full operative detector (full detector installed, final performances in terms of alignment, calibration, etc.)

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 3 SM Higgs boson - Introduction The Higgs boson mass is not theoretically foreseen. Both theoretical and experimental limits exist From direct LEP search: M H > GeV From the Electroweak fit of the standard model M H < 260 GeV M H > 1 TeV is theoretically forbidden The LHC experiments have to cover from the LEP limit up to the TeV scale

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 4 Cross sections… Higgs production cross sections available at NLO. CMS often quote the results with K-factors included. ATLAS usually quote the LO results (preliminary results on NLO). K ggH > 1.7; K ttH ~1.2;K WH ~1.3;K VBF ~1.1 Gluon fusion is the dominant production channel Vector Boson Fusion (VBF) in the following is the next production channel over the whole mass range Though less important in terms of absolute cross section values, associated productions (ttH,WH), provide distinctive signatures. Relevant for largely background dominated decay channels (bb,  ) VBF is characterized by energetic jets in the forward region and by lack of hadronic activity in the central region

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 5 … and Branching Ratios BR bb  WW ZZ LEP excluded  Decays into third generation down-type fermions largely dominating at low mass (M H < 130 GeV) H  bb decay detection feasible only if the associated production ttH is considered. H   suppressed, but it provides a very clear signature. Higgs decay into tau pairs important if VBF production is considered Vector Boson decays imporant at large M H and in the intermediate mass region I will concentrate mainly on the M H < 200 GeV region

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 6 Detector simulation Unless differently stated, the results have been obtained with generator (Pythia, ME) and with a fast simulation of the detector. In fact: Full geant simulation of the detector response More precise detector response Long CPU time for event simulation (O(1 min) per event) Extract detector response from Geant simulation Parametrize the detector response and apply the parametrization to the particle level event Less precise detector response. Short CPU time for event simulation For the most impotant channels, full simulated results have been already obtained. Work in progress for the others

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 7 Higgs decay into bb L = 30 fb -1 k = 1.5  M ~ 15 GeV The region M H <130 GeV is of particular interest (LEP EW fit of the SM) Final state involves 2 jets, 4 b-jets, 1 lepton, E T The lepton provides the trigger. b-tagging and jet energy scale are the crucial detector-related issues. Reducible background: tt+jets,W+jets production Irreducible background: ttbb production CMS applies K ttH =1.5  S/√B = 5.3 ATLAS (no K,ttbb at ME,CTEQ5L)  S/√B = 2.8

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 8 Higgs decay into  100 fb -1 M H =130GeV K=1.6 S/BG ~ 1/20  M : ~1GeV The Higgs decay into  is a challenge for the EM calorimeters. Key points: calorimeter resolution (and linearity), id of photons vs rejection against π 0 CMS(barrel, measured with 5x5 crystal array) ATLAS (barrel, module 0 at TestBeam, η = 0) Main backgrounds: irreducible  dominant.  j and jj together are half the irreducible background Dedicated algorithm for the recovery of photon conversion (~1/3) in the material in front of the EM calorimeter Calorimeter segmentation is critical to reject jets. For ~80% efficiency, the jet rejection factor is 1000 to 4000, depending on the E T

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 9 Higgs decay into ZZ Final states investigated: 2e2μ, 4μ, 4e Irreducible background coming from direct ZZ production Reducible background coming mainly from tt,Zbb The reducible background are strongly reduced by isolation criteria on the leptons and by b-veto The channel is useful for the Higgs discovery in the ranges 130 GeV < M H < 150 GeV and 180 GeV < M H < 600 GeV (H  WW is opened at ~160 GeV)

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 10 Higgs decay into ZZ(2) The H  4μ decay is largely studied in full simulation (ATLAS). Efficiency of the muon spectrometer slightly affected by the layout change after the TDR The studies confirm the results published in the TDR ATLAS p T (GeV) ATLAS

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 11 Vector Boson Fusion During the last years a lot of work has been put on the VBF production channels. Jet One expects two hard jets in the forward and backward regions of the detector  forward jet tagging Lack of color exchange between partons in the initial state  reduced hadronic activity in the central region  central jet veto   Forward tagging jets Higgs Decay Extremely useful for Higgs discovery in the low and intermediate mass region It can be used for several Higgs decay channel  useful to measure Higgs coupling

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 12 Vector Boson Fusion (2) Implementation of fwd jet tagging and jet veto algorithms: Associated jets are “forward”. They are well separated in pseudorapidity Veto any jet in the central region (except decay products of the Higgs) Forward jets are the “signature” of VBF Central jet veto effective for QCD background rejections, in particular against the inclusive tt production, which is a common background for all the VBF channels

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 13 Vector Boson Fusion (3) Detailed tuning of the fast simulation on the full simualtion results has been done. Study undergoing for the new software/new detector layout/new simulation Pileup is a critical issue for the central jet veto. Fake veto rate strongly depends on the amount of pileup and on the veto threshold applied. Results for the new simulation are being produced ATLAS

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 14 VBF H  WW M H =160 GeV WW  e ATLAS 10 fb -1 CMS 60 fb -1 The best results are obtained if the leptonic decay is considered for both the Ws. Dominant backgrounds come from tt production (reduced by the central jet veto and b-jet veto), WW production Selection includes fwd tagging, central and b jet veto, spin correlations between leptons Careful study on the bakground systematics due to generator uncertainties  most of the multijet final states generated with ME generators. There are preliminary results for NLO tt calculation. The channel is one of the most promising for 135 GeV < M H <190 GeV The normalization of the background value can be estimated at 10% level from data. Background shape taken from MC

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 15 VBF H   Investigated both in the llE Tmiss and in the ljE Tmiss channel ATLAS: it provides the cleanest signal for the discovery in the low mass range (M H < GeV) CMS: useful complementary tool to the H   channel The difference relies on the difference calorimeter and muon spectrometer design Main backgrounds: Z+nj, tt, W+nj (hadronic decay of one of the two  ) The Higgs mass can be completely reconstructed using the collinear approximation for the  decay. Defining: It can be shown that

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 16 VBF H   (2) E tmiss (for the ll and lj channel) and hadronic tau reconstruction and resolution (for lj) are key points for the mass measurement E tmiss resolution related to jet measurement and to non-associated clusters in the calorimeters Hadronic tau decay: does the tracker improve the energy resolution? QCD events G4 simulation ATLAS H    μ μ G4 simulation ATLAS PRELIMINARY E T eflow /E T truth 1 prong Z   ATLAS PRELIMINARY

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 17 VBF H   (3) CMS ATLAS 30 fb -1 Statistical significance (Poisson statistics) above 5 for 30 fb -1 if 110 GeV<M H <140 GeV Background normalization control at 10% level from the expected Z   peak and from the high sideband

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 18 Overall view for the SM Higgs LHC is forseen to run for a period (1y?) at reduced luminosity. The discussed results have been obtained assuming L = 2x10 33 cm -2 s -1 (and the corresponding pile-up) The present estimations show that 30 fb -1 of “good” data would be enough for the SM Higgs discovery Above M H = 200 GeV (not shown in the plot) the H  ZZ decay provides a clear, almost background free signature.

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 19 Determination of Higgs parameters Once the Higgs boson would be discovered, we want to measure its parameters (mass, spin, partial widths, coupling constants) Most of the coupling measurements will be possible because in the whole Higss mass range more than one decay is accessible at one time Spin determination examples: spin 1 rouled out if H   or gg  H is observed H  ZZ: spin correlations can be observed in the  φ distribution  --

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 20 Determination of Higgs parameter(2) Precision on the mass: it assumes 0.1% precision on lepton measurement, 1% precision on jet measurement (probably optimistic….) Width determination: detector dominated is M H 200 GeV the H  4l is used for direct determination.

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 21 Ratios of Higgs decay widths and couplings Performed a maximum likelihood fit to all channels, taking into account cross talk between signal channels, systematics (luminosity, reconstruction efficiencies, backgrounds) Assuming that only SM particles couples to the Higgs, the ratios between coupling constants can be extracted. With additional hypothesis (couplings to W and Z as in the SM, no new particles enter in the loops for the  decay) the absolute coupling can be determined

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 22 MSSM Minimal Supersymmetric extension: two Higgs doublets  8 degrees of freedom (5 particles): CP-even : h,H CP-odd: A Charged: H +,H - Couplings to SM particles modified w.r.t. SM. Decay into third generation fermions enhanced at high tgβ At high M A the heavy bosons degenerate in mass while the h saturate at a limit value (around 130 GeV)

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 23 h search at LHC Most of the studies done in the LEP maximal mixing scenario (μ=-200 GeV, M SUSY = 1 TeV, M 2 = 200 GeV, M g = 800 GeV) In the decoupling limit (high M A ), the h behaves like a SM Higgs boson. Decay into bb enhanced at high tgβ h  ,h  bb only Most of the parameter plane covered by h  bb h  μμ important at high tgβ at low M A VBF channels not included in the plot

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 24 Heavy Neutral Higgs Bosons bbH/A  bbμμ: cross section higher than in the SM at high tgβ. Studied performed in full simulation by CMS. Main backgrounds coming from Drell-Yan production and tt. At 130 GeV, the resolution (1%) is not enough to resolve the mass difference between H and A Most of the background rejected with the requirement of b-tag. Since the associated b are soft (E T < 50 GeV), a special tagging algorithm (use of secondary vertex and impact parameter without the jet requirement in the calorimeter) has been used to increase the b-jet ID

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 25 H/A decay into  ’s The tau decays provides the cleanest signature for the havy Higgs discovery at high mass (and relatively high tgβ) Investigated all the final states (ll, lj, jj). All of them contribute (at different M A ). The associated production (bbA/H) provides additional rejection against the main backgrouds (Z+jet, WW, QCD (double hadronic tau decay only) Trigger issues solved w.r.t. the full hadronic final state

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 26 Overall coverage of the M A -tgβ plane for A/H The A/H Higgs boson can be detected thanks to the A/H  μμ decay in the low mass- high tgβ region The decay into  allows (with different final states) to cover a large part of the plane There is no coverage (with 30 fb -1 ) for the large mass- intermediate tgβ region The very high mass region can be reached only by the full hadronic final state of the  Similar plot for ATLAS

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 27 H +   ;   hadr Reconstruction of the top decay and of the hadronic tau. tt and Wt backgrounds suppressed exploiting spin correlations  One single π has to carry the largest part of the tau energy. large background (tt+jets) reconstruct tops and Higgs  M H H +  tb t  bW  blnqq CMS L= 30 fb -1 Charged Higgs decay Two channel are of particular interest if M H > M t.

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 28 Overall view (10 fb -1 ) No VBFwith VBF Few statistics (of good data) is enough to cover most of the M A - tgβ plane But a lot of questions to address before: machine operations, detector understanding, background measurement….

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 29 Overall view (300 fb -1 ) Can we disentangle between SM and MSSM just from the Higgs sector? In most of the parameters space more than 1 Higgs boson is detectable. In the yellow region just the h is detectable Still in a region we can disentangle looking at the ratio between the BR into  and WW. Anyway, it is difficult….

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 30 Conclusions -Standard Model Higgs detectable with few tens of fb -1 over all the mass range. In the most likely region, VBF production plays a relevant role -Spin measurement feasible in its ZZ decay. -Coupling measurement will require full luminosity in most of the mass range. Achievable precisions: 15-50% depending on the channel -MSSM plane coverage requires few tens of fb -1 in a maximal mixing scenario. Higgs decays into third generation fermions play an extremely relevant role. VBF important for the h in the decupling limit

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 31 BACKUP

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 32 Signal(mH=130GeV) ZZ  4  ZZ  2  2  ttbar Zbb Results for L=30 fb -1 3) Results for L=30 fb -1 Signal (mH=130GeV) (eff=81.2%)4.16 ZZ  4  1.36 ZZ  2  2  0.15 Zbbar  4  0.31 ttbar  4  0.01 BKG1.83 Signal and background rates after overall analysis in mass window ±5 GeV around mH Higgs Mass (GeV) Significance (Poisson) By combining the channels H  ZZ*  4 , H  ZZ*  4e, H  ZZ*  2e2 , the Higgs signal can be observed with a better than 5σ significance over most of the range 130<m H <180 GeV for an integrated luminosity of 30 fb -1 Before after

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 33 M T >175 GeV M T <175 GeV Signal Region Outside Signal Region

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 34 2) Analyse: Background rejection  Signal and backgrounds after kinematic cuts ( integrated over a mass window of ± 5 GeV around m H =130 GeV) Rejection of irreducible background 2) Rejection of irreducible background Use Likelihood function & Neural Network with discriminating variables based on Higgs properties Aim : Bring the reducible bkg down to ~ 10% of irreducible bkg (protection vs theoretical uncertainties)  Rejection ~ 100 is needed processσXBR(fb) Signal : H  4 µ (m H =130 GeV)0.10 Irreducible : (Z/  *)(Z*/  *)  4 µ0.04 Irreducible : (Z/  *)(Z*/  *)  2µ2 0.01 Reducible : (Z/  *)bb0.40 Reducible : tt0.27 Rejection of reducible background 1) Rejection of reducible background

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 35

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 36

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 37

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 38 What’s new for Muon System since PhysicsTDR (1999)? Provide access to EndCap Calorimeter and ID Central crack ~ 45 m ~ 25 m

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 39

WIN 05 -Delphi 10/6/2005 Iacopo Vivarelli-INFN Pisa 40 Masses and tan from neutral Higgs bosons Error on masses m/m = 0.1 to few % Error on tan  tan/tan = 15 to 5 % from rates of H/A   VBF h/H   not studied yet ATLAS TDR