April 18th, 2009 TILC09 - Corrado Gatto 1 Higgs Studies in the 4 th Concept Corrado Gatto INFN...

Higgs Studiesin the4th Concept Corrado GattoINFN Napoli/LecceOn behalf of the 4th Concept Collaboration

TILC09 - Corrado Gatto

4th Concept Software Framework: ILCrootCERN architecture (based on Alices Aliroot)Full support provided by Brun, Carminati, Ferrari, et al.Uses ROOT as infrastructureAll ROOT tools are available (I/O, graphics, PROOF, data structure, etc)Extremely large community of users/developersTGenerator for events generationVirtual Geometry Modeler (VGM) for geometry Virtual Montecarlo for particle transportGrowing number of experiments have adopted it: Alice (LHC), Opera (LNGS), (Meg), CMB (GSI), Panda(GSI), 4th Concept, LHeC and the forthcoming International Dual Readout Collaboration Six MDC have proven robustness, reliability and portabilityDo not Reinvent the wheelConcentrate on Detector studies and Physics


From Bangalore to Now:Detectors in ILCroot4th Concept BaselineVTX: from SiD scaled to 3.5 Tesla (original version)Drift Chamber: 2nd versionFiber Triple Readout Calorimeter: 3 versionMuon Spectrometer: Original versionNew additions:Crystal Triple Readout Calorimeter Included into the LoIFTD (from SiLC) Did not make it into the LoIAlso availableVTX Detectors: Original SiD Central Trackers: TPC, Si-Strips (SID01), SPT (Pixel Tracker)

Total: 10 subdetectors (16 versions), most of them with full simulation


The Virtual Montecarlo ConceptVirtual MC provides a virtual interface to Monte CarloIt allows to run the same user application with all supported Monte Carlo programsThe concrete Monte Carlo (Geant3, Geant4, Fluka) is selected and loaded at run timeCompare Montecarlo performance and possible flows Choose the optimal Montecarlo for the study


Physics Studies in the 4th ConceptDetector simulation frozen in July 2008 (except Ecal). Simu & Reco started August 2008Several event samples used:4th Concept (Pandora-Pythia, Whizard, Evtgen, etc,..)ILD sample for LOI (many thanks to Frank and Akiya)KEK sampleFluka and Geant4 (through VMC)Not only ILCroot: MarlinKinFit & Rave99% computing resources are from FermilabILCroot is freely available at Fermilabhttp://ilc.fnal.gov/detector/rd/physics/technical/resources/ilcroot.shtml


Processing FlowVMCSteerReconstructionLocal tasksGlobal tasksAll detectors are fully simulated (except gaussian smearing of hits in the Muon Spectrometer):Local tasksLocal tasks


Recoil massanalyses


e+e- -> ZoHo -> m+ m- X s=250 GeVOlder AnalysisKEK event sample for signal and SM backgroundISR and beam spread (0.25%) included-80% e- polarization, +30% e+ polarizationAdd GuineaPig events at digitization step (1BX for DCH and 10BX for VTX)Geant4 for particle transport of signal eventsGeant4 for particle transport of beam background eventsNo calorimetry used

signalbkgndDCH Occupancy

Final Statee+e- -> HZ; H->X, Z->mme+e- -> ZZ; Z->X, Z->mm


e+e- -> ZoHo -> m+ m- X s=250 GeVOlder Analysis

Results


e+e- -> ZoHo -> m+ m- X s=250 GeVfrom LOI

signalbkgndILD event sample for signal and SM backgroundISR and beam spread (0.3%) included+100% e- polarization, -100% e+ polarizationAdd GuineaPig events at digitization step (1BX for DCH and 10DX for VTX)Fluka for particle transport of signal eventsGeant4 for particle transport of beam background eventsECAL+HCAL with implementation of Triple Readout

Worst caseScenarioLarge W+W- background


e+e- -> ZoHo -> m+ m- X s=250 GeVAnalysis strategyInitial cuts to reduce background

Require two tracks in the Muon Spectrometer

G. Tassielli, G. Terracciano, M. Peccarisim+m- efficiency vs PSingle m- efficiency vs P


e+e- -> ZoHo -> m+ m- X s=250 GeVResolution of Muon Spectrometer

G. Tassielli, G. Terracciano, M. Peccarisi


e+e- -> ZoHo -> m+ m- X s=250 GeVResults

G. Tassielli, G. Terracciano, M. PeccarisiFinal cuts for S/N enhancement


e+e- -> ZoHo -> m+ m- X s=250 GeVResolutions

G. Tassielli, G. Terracciano, M. PeccarisiResults affected by Mm=0(and consequentm brehemstralung)

Recoil MassFromMC Truth


e+e- -> ZoHo -> e+ e- X s=250 GeVfrom LOI

signalbkgndILD event sample for signal and SM backgroundISR and beam spread (0.3%) included+100% e- polarization, -100% e+ polarizationAdd GuineaPig events at digitization step (1BX for DCH and 10DX for VTX)Fluka for particle transport of signal eventsGeant4 for particle transport of beam background eventsECAL+HCAL with implementation of Triple Readout


e+e- -> ZoHo -> e+ e- X s=250 GeVECAL is crucial in this analysis

G. Tassielli, G. Terracciano, M. Peccarisie- energy resolution from VTX+DCHe- energy resolution from VTX+DCH+ECAL


Particle Identification with Triple Readout45 GeV particlesepmepmILCRoot simulationNew resultswith neutrons contribution


e+e- -> ZoHo -> e+ e- X s=250 GeVAnalysis strategyInitial cuts to reduce background

Require two electrons in the ECAL/DCH

Final cuts for S/N enjhancement

G. Tassielli, G. Terracciano, M. Peccarisie+e- efficiency vs Eeee = 93.4%Purity= 98.2% e+e- efficiency vs qeee = 93.4%Purity= 98.2%


e+e- -> ZoHo -> e+ e- X s=250 GeVusing VTX + DCHResults

G. Tassielli, G. Terracciano, M. PeccarisiRecoil MassRecoil mass resolutione+e- mass resolution


e+e- -> ZoHo -> e+ e- X s=250 GeV using VTX + DCH + ECALResults

G. Tassielli, G. Terracciano, M. PeccarisiRecoil MassNeed a better integration of the informations from tracking and calorimetryRecoil mass resolutione+e- mass resolutionLong tails disappeard

Mass shiftappears


Jet analyses

ILD event sample for signal and SM backgroundKEK for WWnn and ZZnn analysisISR and beam spread (0.3%) included+100% e- polarization, -100% e+ polarizationNo beam backgroundFluka for particle transport of signal eventsNo ECAL


Jet reconstrucion:combine calorimetric and tracking informations

(work in progress)

A. Mazzacane


Jet Reconstruction Strategy


Jet Reconstruction StrategyJet axis 1Jet axis 2Use informations from tracking Systems only


Jet Reconstruction StrategyCone 1Cone 2Use informations from tracking calorimetry onlyWith recursive ycut


Jet Reconstruction StrategyOutlierJet axis 2OutlierOutlierOutlierUse combinedinformations from tracking and calorimetry


Jet Reconstruction StrategyMuonInclude muon spectrometer


e+e- -> ZoHo -> nn X s=250 GeV from LOI

ILD event sample for signal and SM backgroundISR and beam spread (0.3%) includedNo Beam backgroundFluka for particle transport of signal eventsOnly HCAL with implementation of Triple Readout (ECAL implemented later)

signal2 jetsbkgnd4 jetsbkgnd

Final Statee+e- -> HZ; H->qq, Z->nne+e- -> HZ; H->qq, Z->qqe+e- -> W+W- ->qq lne+e- -> ZZ ->qq l l


e+e- -> ZoHo -> nn X s=250 GeVA. MazzacaneDisentangle 2-jets from multi-jets events

Reduce ZZ and WW background

Reject prompt muons (via the Muon Spectrometer) from leptonic WW decay

e+e- -> HZ ->qq nn+e+e- -> ZZ ->qq nn

Already studied


e+e- -> ZoHo -> nn X s=250 GeV Results

Next step is to consider e+e- -> ZZ ->qq nnande+e- -> ZZ ->qq qqe+e- -> W+W- ->qq qq

signalA. MazzacaneWWZZRave Flavour taggingccbbudsRecoil Mass(no 4-jet survivors)


e+e- -> ZoHo -> nn cc + ZZ BackgroundSignal + ZZ backgroundRequires 2 jets from jet-finderEvis > 130 GeV

No flavor taggingFit with three gaussians Selection Efficiency = 80.2%1% Error on the massTriple ReadoutCompare with our previous analysisUsingKEK data (thanks to A. Myamoto)


e+e- -> ZoHo -> 4 jets analyses

(work in progress)


e+e- -> ZoHo ; and Zo->cc Ho->bb s=250 GeVF. IgnatovAnalysis strategySelect Event with 4 jets (use jet finder with recursive ycut)Ecalo + E muon cut to reduce background(events with neutrino or ISR)5-C kinematic fit to all possible jet-jet combinations

Pick combination with highest probabilityFinal cut: c2/ndf

e+e- -> ZoHo ; Zo->uu Ho->cc s=250 GeVY. C. ChenAnalysis strategySelect Event with 4 jets (use jet finder with recursive ycut)Select Mj1j3 and Mj2j4Requires 1 combination within 10 GeV from nominal Z massPlot the other combination

Signal onlyNo BackgroundEvents with significant Emiss (neutrinos)Analysis is still in progress


W/Z Mass Separation KEK event sampleSimple Durham jet-finder a la L3 (recursive ycut) used for this analysis No combined information with tracking yet (3 entries/evt)No ECAL4-jets finding efficiency: 95%A. Mazzacane


Status and PerspectivesILCroot framework is stable and very fastIt runs smoothly on the GRID at FNALVery detailed simulation & reconstruction takes 20-60/eventMost of the effort by the 4th Concept Simulation group has been put into implementing the baseline detectors + several optionsSome work on the detectors is still needed (detector optimization and SILC+CLUCOU)Lots of Physics studies will followAdd other background sourcesAnybody is welcome on board


Backup slides


Event Display in ILCrootLow pt secondary muone+e- -> HoHoZo -> 4 jets 2 muonsECM = 500 GeV


OutlineThe simulations in the 4th ConceptILCrootDetector simulations Performance & OptimizationPhysics benchmarks for the LoIFuture prospects


The Virtual Montecarlo ConceptVirtual MC provides a virtual interface to Monte CarloIt allows to run the same user application with all supported Monte Carlo programsThe concrete Monte Carlo (Geant3, Geant4, Fluka) is selected and loaded at run timeCompare Montecarlo performance and possible flows Choose the optimal Montecarlo for the study


4th Concept Software Strategy: ILCrootCERN architecture (based on Alices Aliroot)Full support provided by Brun, Carminati, Ferrari, et al.Uses ROOT as infrastructureAll ROOT tools are available (I/O, graphics, PROOF, data structure, etc)Extremely large community of users/developersTGenerator for events generationVirtual Geometry Modeler (VGM) for geometry Based on Virtual MontecarloCould it ever evolve into a general purpose entity for the HEP community (as ROOT)?Growing number of experiments have adopted it: Alice, Opera, CMB, (Meg), Panda, 4th ConceptSix MDC have proven robustness, reliability and portabilityDo not Reinvent the wheelConcentrate on Detector studies and Physics


Dual Readout CalorimetryTotal calorimeter energy: use two measured signals and two, energy-independent, calibration constantsFrom calibration@ 1 Energy only Non-gaussianNon-gaussiangaussian40 GeV p


Improving the Energy Resolution: The Effect of Neutrons45 GeV -ILCRoot simulation


From Dual to Triple ReadoutInclude the effect of neutrons45 GeV -ILCRoot simulationTriple readout aka Dual Readout with time history readout


Compensation withECAL and HCALGet EScint and ECer from ECAL (disregard neutrons as ZBGO >> 1)Get EScint, ECer and Eneutr from HCALThen:

Estimate hC, hS and hneu from a 45 GeV run (p- and e-) by minimizing the spread of Etot


Calorimeter Response for 45 GeV e-


80 GeV jet with escaping particlesILCRoot simulation


m+ m at 3.5 GeV/cILCRoot simulation


Jet reconstrucion:combine calorimetric and tracking informations

(work in progress)


Jet Reconstruction StrategyMuon


SummaryWorst case polarization scenario considered: largest WW background


April 18th, 2009 TILC09 - Corrado Gatto 1 Higgs Studies in the 4 th Concept Corrado Gatto INFN...

Documents

Transcript of April 18th, 2009 TILC09 - Corrado Gatto 1 Higgs Studies in the 4 th Concept Corrado Gatto INFN...