PHOTON RECONSTRUCTION IN CMS Jukka Nysten Helsinki Institute

PHOTON RECONSTRUCTION IN CMS Jukka Nysten Helsinki Institute of Physics for the CMS Collaboration Jukka Nysten ACAT 03 1

Outline • Accelerator and experiment • Online event selection: level 1 and high level trigger (HLT) • H gg channel • Photon conversions • Offline analysis: handling of unconverted and converted photons Jukka Nysten ACAT 03 2

Equipment • LHC – – – • Proton proton collider Collision energy 14 Te. V Luminosity Bunch crossings every 25 ns 20 interactions in each bunch crossing CMS – General purpose experiment – Silicon tracker: inner pixels, silicon microstrips – Crystal electromagnetic calorimeter: lead tungstate crystals (22 mm x 22 mm) • Simulations – Event generation: Pythia – Detector simulation: CMSIM (Geant 3), OSCAR (Geant 4) – Reconstruction and electronics simulation: ORCA Jukka Nysten ACAT 03 r z 3

Triggers: Level-1 & HLT • Bring event rate down – Select signal events with high efficiency – Reject background events with many orders of magnitude L 1: 40 MHz 100 k. Hz HLT: 100 k. Hz 100 Hz Level-1 HLT Level-2 ECAL reconstruction Threshold cut Level-2. 5 Pixel matching • Hierarchical structure Level -1: hardware HLT: software Jukka Nysten Photons Threshold cut Isolation ACAT 03 Level-3 Electrons Track reconstruction E/p, matching (Dh) cut 4

HLT output rates Jukka Nysten ACAT 03 5

H gg signal extraction • Promising signal for Higgs boson mass region m 2 < 150 Ge. V/c H • Large background – Reducible background (neutral pions from jets) – Irreducible background (prompt photons) • An excellent energy resolution and the knowledge of the primary vertex is required Jukka Nysten ACAT 03 6

Photons in the tracker • Large amount of material in the tracker volume • Photon can convert into an electron positron pair in the tracker material with high probability • Probability of a photon converting: • Energy deposit in ECAL is spread because the 4 T magnetic field affects the • 70% of H gg events charged particles have at least one degradation of energy converted photon resolution Jukka Nysten ACAT 03 7

Offline photon reconstruction • • Energy measurement Conversion handling Primary vertex finding p 0 rejection Jukka Nysten ACAT 03 8

Energy measurement ECAL clustering algorithms: • Fixed window: for unconverted photons • Hybrid: barrel • Island: endcaps Jukka Nysten ACAT 03 9

Energy reconstruction for converted photons Separation of e+ e- pair on ECAL Photon energy resolution unconverted Jukka Nysten Use different algorithms for energy reconstruction depending on the conversion pattern. Converted photons have worse energy resolution than unconverted ones ECAL noise terms converted ACAT 03 10

Track finding for conversions • Use the Kalman filter for track reconstruction • Track inward in the tracker barrel • Form the initial guess from the super-cluster and the hit on the outermost tracker layer Jukka Nysten Conversion tracking should provide: – conversion position – z coordinate for primary vertex – means to do p 0 rejection ACAT 03 11

Z coordinate of the primary vertex • • • Track extrapolated to the beam line in the (r, z) plane Data: photons with no pile up, tracker barrel region Stereo layers essential for position resolution (r=20, 30, 60, 70 cm) r < 20 cm Jukka Nysten r 20 cm< r < 65 cm ACAT 03 z r > 65 cm 12

Primary vertex • Procedure for finding the primary vertex with charged tracks: – Reconstruct charged particle tracks in the event – Reconstruct vertices – Choose the Higgs production vertex • Uses the knowledge that Higgs production events are harder than pileup events. • At high luminosity choosing the right vertex is quite difficult. • Primary vertex location from a converted photon might help. Jukka Nysten ACAT 03 13

Summary • Precise photon reconstruction is required to discover the Higgs boson in the two photon channel. • Photon selection is done in level-1 and High level triggers. • Unconverted and converted photons require separate algorithms. • Reconstructing conversions can provide information about the z coordinate of the primary vertex, conversion radius and help in p 0 rejection. Jukka Nysten ACAT 03 14

References • The Electromagnetic Calorimeter Project Technical Design Report CERN/LHCC 97 -33 CMS TDR 4 15 December 1997 • CMS Technical Proposal CERN/LHVV 94 -38 (1994) • The Tri. DAS Project Technical Design Report, volume 2. Data Aquisition and High-Level Trigger CERN/LHCC 02 -26 CMS TDR 6 15 December 2002 • E. Meschi, T. Monteiro, C. Seez, P. Vikas, CMS Note 2001/034 Jukka Nysten ACAT 03 15